Hockey One, Hockey Two

On the other thread we are starting a discussion on the warming periods over the last few millennia, and so I thought I put up the second of the youtube clips that deal with it. Note that the “Medieval Warming” was a local event, and that when global details are taken the warming disappears – at least that is what the US National Academy of Science seems to say. They were “critical of the process” but did confirm the findings.

Plimer has used the graph without acknowledging (to my knowledge as I am relying on others who have read the book) that the basic details of the “hockey graph” are actually supported by the evidence, whereas his graph showing the “medieval warming” has been discredited.

Advertisements

358 Responses

  1. Neil and Sparta

    Hope you guys understand that I was not trying to shut down debate on the other thread, just that we can bring it onto here.

  2. Joni,

    Of course not mate…….We may not agree on much but you have never tried to “shut down” anything that I am aware of………Cheers…….

  3. Both those graphs are in Plimers book. He says the graph with the big hump is the correct one and that the hockey stick graph is a fraud.

    Big claim for Plimer to make

  4. Thanks Neil/Sparta – I do not want to shut down debate, debate is what helps us learn,

    But the hockey stick one has been refined by those who produced it after the criticism, and it is still basically correct. Whereas the medieval one was not updated with the new data. So which one is misrepresentation.

    And sleep now for me. Nite.

  5. Sparta of Phoenix, AZ USA, on May 25th, 2009 at 12:36 am Said:

    I am sorry but this “straw-man” excuse is getting very old already.

    It is always being bought up because far being an excuse the opponents and right keep brining up straw-men. It’s an sop for them.

    The number of “scientists” somehow adds credence to the argument; all with paychecks now dependent on that “support”? I am sure it has nothing to do with grant funding or anything, it is all in the interest of mankind……

    This is what is getting very old that false claim, and the claim it’s all one big global scam or a zealous religious movement.

    What about that Vostok ice core data anyway?

    http://www.ncdc.noaa.gov/paleo/icecore/antarctica/vostok/vostok.html

    Hope you have a lot of time to read all those references.

    Hmm…….but I thought the debate was over?

    Who said the debate was over, certainly not the scientists who are researching this and that is where the real debate lies and should lie? What we are waffling on about here and others are doing on other blogs is out of ignorance and mostly along ideological lines.

    And aren’t we debating here so where is the debate over and just who declared it over, oh you did in being facetious, which is your usual want.

  6. johnd, on May 24th, 2009 at 9:27 pm

    Picking up one off isolated events (and even having a good laugh over it) to make a point.

    The BOM got it wrong so according to you it’s all wrong. But what of all the times the BOM got it right and saved farmers and industry money, something they continue to do to this day?

    But no you rather they just shut down or become an ostrich and hope nothing bad happens. For mine, and I’d like to believe for a lot of rural people, I would rather the BOM make the predictions and be very wrong every now and again than just bury their head in the sand for fear of getting it wrong.

  7. Neil,

    Carried over from the last thread, Howard actually took an ETS policy to the election as well.

  8. Nice video, Joni, but, not surprisingly, I’m unconvinced.

    Anyone wishing to pad out their knowledge on the subject could do worse than read this presentation to the Ohio State University by Steve McIntyre, the Canadian statistician who originally challenged the validity of the hockey stick graph.

  9. “What we are waffling on about here and others are doing on other blogs is out of ignorance and mostly along ideological lines.”

    Well, I am quite impressed you would actually admit ignorance but alas, you are reborn. You raise a point though. With so many still in “ignorance”, to include our political leaders, why would you and they propose/expect the average “Joe” to accept economic hardship in the name of plausible ignorance? Why are you or they so sure exactly? The computer model says so? Your link by the way does nothing to correlate the “physical evidence” with the “computer models”….In fact, if you bothered to read my link, which is one of many, you might have noticed the rather plain to see contradiction…The core date implies it is increasing temperatures that raise CO2, not the other way around…….

    “And aren’t we debating here so where is the debate over and just who declared it over, oh you did in being facetious, which is your usual want.”

    Not really Adrian, we have some who purport to know for sure what is happening or at least to the extent that we need to make economically harmful changes and those of us not yet convinced for lack of evidence surrounding the most basic of questions. Meanwhile millions in grants, burgeoning industries/schemes etc are being proposed based on an unfinished debate, which is routinely passed off as “fact”.

    Why you believe this is not odd is not hard to imagine. Facetious at times I may be, but for somebody that recently criticized me for not understanding enough about the Aboriginal movement to even mention the “apology” only to condemn a source you had not even read yet says much about you I am afraid…..LOL….I dare say what your “usual want” smacks of……

  10. With all due respect Sparta, governments frequently ask us to suffer “economic hardship” in ignorance of the exact data, theories, & conclusions they have in front of them.

    For example, your much vaunted war in Iraq was based on the supposition that Saddam Hussein still had WMD’s. It was a theory with flimsy evidence but, let’s assume for the sake of argument the majority of intelligence & models based on said in intelligence was aligned with the fact he had them. This is the same position in the scientific community in regards to man-made global warming – the majority believe it to be true.

    Now, you still support the war in Iraq, even thought the initial premise was to prevent the WMD’s Saddam Hussein “had” being given to Al Qaeda (again assuming for sake of this debate that this was feasible & Saddam had a semi-working relationship with them). WMD’s have not been found, but the public (in the US at least) is being asked to fork out billions upon billions to keep this up.

    If you are so bothered by the public purse being used for theories that may not be proven out – why are you not up in arms in regards to the war in Iraq? I am not interested in a debate on Iraq & the War on Terror mind you – just bringing up a parallel in which huge economic hardship is occurring today based on flawed evidence &/or flawed theories.

  11. Mobius Ecko, on May 25th, 2009 at 7:38 am Said:

    The BOM got it wrong so according to you it’s all wrong. But what of all the times the BOM got it right and saved farmers and industry money, something they continue to do to this day?

    Mobius, weather forecasts are probably only a topic of conversation for you. If your livelihood relied on accurate forecasts then you might have a different opinion. BOM’s strike rate is not very high, in fact it’s unacceptably low for many people. In recent years they have started issuing forecasts such as “in the next period there is a 50% chance of above average rain and a 50% chance of below average rain”. Go on, laugh, but that is no joke, that is a real forecast.
    That is why many people, particularly those in agriculture turn to private forecasting services, typically paying $600 annually for a basic service. A typical private forecaster is someone who previously worked with BOM but left disillusioned with their rigid and narrow focus, failing to keep up with advances. Perhaps the most telling, and most important for people in southern and central Australia is their tardiness in incorporating Indian Ocean data into their modelling.
    This failure has been beneficial for those private forecasters that do incorporate it and are perhaps decades ahead of BOM in their knowledge of how Australia’s weather really works.
    $600 is really cheap insurance, users of such services tell of saving or gaining amounts of up to $100,000 through having access to forecasts they know have a very high degree of accuracy and that they can rely on with confidence.
    In the example I gave, BOM didn’t get it wrong because they can’t get everything right, they got it wrong because the models on which they make all their forecasts are flawed by failing to give weight to perhaps the most significant driver of weather over a large part of Australia.

    Sometimes you have to work backwards to get the real answers. For instance BOM will tell you how many El-Nino years were drought years, inferring that an accurate forecast of an El-Nino event will be an accurate guide to the chance of a drought. However look at it the other way and determine how many drought years were El-Nino years, you’ll get a very different answer, and that then gives an indication of why BOM’s accuracy is unacceptable to many people.

    So, Mobius tell me about the times BOM got it right.

  12. “Now, you still support the war in Iraq, ……..”

    Well although this is nothing more than a fruitless distraction, most of us didn’t simply think it was simply over WMD’s despite the continued rhetoric from your ilk; a major component indeed but not the only reason for many of us. One more time for the record, 16 resolutions, repeated firing on our aircraft, blah, blah, blah….. We didn’t rebuild German, Japan, or Italy as a part of a major grand strategy until after the fact either but a successes they were…..Wars evolve mate, every one of them has, get over it. Notice nobody is interested in talking about it now that things have gotten better, pathetic….. unlike AGW we certainly have an ending in sight now don’t we. When pray tell will we have the global warming boogey man licked exactly? Oh yes, when we reduce the cause we aren’t yet certain is the cause, brilliant? After we have wrecked our economies while China and India press on, nope, not even then….When exactly is the end?

    “If you are so bothered by the public purse being used for theories that may not be proven out – why are you not up in arms in regards to the war in Iraq?”

    Umm…..the Iraq war is won (a war many like yourself said couldn’t be one may I remind you) and we will be all but gone in a year or two (for the most part). The Iraqi’s can do with their freedom as they will. The question of WMD is answered after all, now isn’t it? Why no mention of Afghanistan, oh yes, the good war waged because why again? Oh yes, that same intelligence you mock that suggested Bin Laden was there, good reason to invade a country. Whatever the cost to the “public purse” we still have a result that can be measured in the “real world” now don’t we. Can the AGW proponents say as much even in our own lifetime? We don’t even know what were fighting at this point for heaven’s sake but you say we should spend and tax anyway? Marvelous, very GW-like of you….LOL…..

    “I am not interested in a debate on Iraq & the War on Terror mind you – just bringing up a parallel in which huge economic hardship is occurring today based on flawed evidence &/or flawed theories.”

    Then don’t bring it up to try and make a point. Iraq neither is hardly an untested theory nor flawed unless you would consider Germany and Japan as such….LOL……..

  13. Mobius, I just have to share this with you.
    Last year BOM issued a forecast “there is a 40% chance of above average rain over the next 3 months”. One newspaper then wrote a front page article about the good news for the farmers, above average rains are coming. When I contacted the journalist it took some talking to convince her that she had gotten it wrong. As far as she was concerned BOM had forecast above average rains, even if there was only a 40% chance.
    I don’t think it was entirely her fault, BOM by using the words “above average” had deliberately put a positive spin on their forecast. If their forecast had been more correctly “there is a 60% chance of below average rain over the next 3 months” nobody could have gotten it wrong. Except BOM of course.

  14. Sparta of Phoenix, AZ USA, on May 25th, 2009 at 9:20 am Said:

    “Now, you still support the war in Iraq, ……..”

    Sparta, your mention of Germany and Japan at the end of your post is somewhat relevant to what seems to be the rewriting of historical temperature reconstructions. Rewriting history, either directly or by omission is a subject both nations are familiar with. If whole nations can attempt, it must be a piece of cake for a relatively few scientists, especially as there is no-one still alive from the era in question to expose them.

  15. @Sparta:
    Seriously missed the point didn’t you? The issue is not about the Iraq war – a point I mentioned and even gave you the benefit of contested facts on. The point is that governments, including the US & Australia, made huge economic & personnel expenditures based on faulty evidence & theories. By your own words, the WMD’s – which do not exist – were a “major component”. As far as selling the war to the public, they were the primary component, but even as a “major” component – my point still stands.

    That is, your use of “economic hardship” is a red herring, as you are more than willing to forgive larger economic expenditure on unproven claims provided they meet your personal world view.

    As for winning the war – given 85 members of the US military personnel have died in Iraq since Obama was inaugurated… I wouldn’t say the war is won yet. But I’m going to let the inevitable reply to that slide because it is irrelevant to subject at hand.

    @johnd:
    Again with the conspiracy stuff!?! If we cannot trust the data put before us, neither side can win this argument.

  16. johnd, on May 25th, 2009 at 9:43 am

    Do your observations apply to a scientist like Plimer who is, in part, relying on an interpretation of the geological record to extrapolate one version of a Gaussian future, johnd? It must be kind of hard finding scientists from either of those past or future eras to expose when writing an inductive history, not unlike Sparta’s backwards and forwards inductions which omit Korea, Vietnam, Afghanistan (one in the same with Iraq IIRC), and indeed the verity of Baghdad where the bombs are still going off in the streets. 😉

  17. How Bolt twists the facts, especially on climate change.

    Arctic Sea Ice Extent

    When the 2009 blue line was trending toward the grey Average earlier this year Bolt used this graph to prove how all the proponents and especially the scientists have got it so wrong. That one graph was the proof that global warming is a myth.

    Since the 2009 blue line has trended away from the average, quite dramatically, Bolt refuses to acknowledge this data, data not so long ago he was using as irrefutable evidence when it appeared to support his denialist stance.

    Plimer does a similar thing. Uses data from periods that support his contention but deliberately leaves out or ignores updates to the same data sets that doesn’t. It’s worse for Plimer because though Bolt likes to think he knows all the science, and even at times more on a speciality than the scientists, Plimer is a scientist and knows just how wrong what he has done is.

    Footnote:
    It is well worth a good look at the NSIDC website just for the fascinating subject it is: http://www.nsidc.org/

  18. Plimer does a similar thing. Uses data from periods that support his contention but deliberately leaves out or ignores updates to the same data sets that doesn’t.

    Still giving ‘expert’ comments on a book you haven’t read? Amusing, to say the least.

  19. @Adrian / Mobius Ecko:
    Tony has a point here. A minor one given the coverage the book has gotten for it’s inaccuracies & bunkum science, but a point nonetheless.

    Get a copy of the book (I’ll send you one if interested – bug me offline), then Tony’s misdirections on the subject can cease.

  20. Personally I’ve decided to move past the “is it real or not?” discussion about climate change.

    The most responsible position is to encourage a greater sense of care about the environment and some frugality about consumption.

    Our severe economic decline will probably reduce carbon producing economic activity more that a carbon trading scheme in any event.

    The really annoying pint in the current debate it the stupidity of the focus on carbon trading, and using a price mechanism as the primary means of controlling emission.

    Imagine the justifiable outrage if this practise was applied to water consumption. Heavy price increases during periods of drought, and low prices otherwise. Would anyone seriously suggest that this would not impose hardship, and would produce less beneficial outcomes than other mechanism to control demand?

    We are placing so much emphasis on the price mechanism, and very limited emphasis on the types of measures that have successfully been applied in reducing excess water consumption.

    Here are some lifestyle/consumption questions/issues that I’d like to know about –

    • Does beef consumption produce more carbon that lamb, chicken, kangaroo? How much?
    • How much carbon is produced in the manufacture of a motor vehicle? How much carbon would be saved if the government provided the tax benefit to 5 years?
    • Why isn’t every consumer item graded on the carbon used in manufacture? Why is there not some “good practice” standard?
    • Why not mandate heating and cooling thresholds for homes and offices? – eg homes and offices not to be heated above 18 degrees or cooled below 24.

    In the absence of policies that directly educate people, I think the carbon trading scheme is a complete cop out.

    Thank you for listening.

  21. “That is, your use of “economic hardship” is a red herring, as you are more than willing to forgive larger economic expenditure on unproven claims provided they meet your personal world view.”

    What tax has been laid upon the American people to fight the war? Red herring, are you really that daft? It was your analogy dude; my apologies it was so easily dismantled?

    “I wouldn’t say the war is won yet. But I’m going to let the inevitable reply to that slide because it is irrelevant to subject at hand.”

    Well if your measure is the complete end to violence than you are dreaming, it is the Middle East after all………

  22. B.Tolputt, on May 25th, 2009 at 10:21 am Said:
    @johnd:
    Again with the conspiracy stuff!?! If we cannot trust the data put before us, neither side can win this argument.

    I’m not sure what you think is conspiracy stuff, but I don’t think that we should ever blindly accept information put before us, by anyone.
    You don’t have to be a scientist or an academic to run certain tests on it, such as does it make sense, is it consistent, or does it meet the test of logic.
    Not that logic is always apparent, at times it requires a bit of lateral thought to see the logic, an oxymoron perhaps.

  23. Tom,

    I can agree with your theme but how do we even know at this point that it is CO2 we should be concerned about? Why do you think I go on about population dynamics in regards to your theme?

  24. B. Tolputt,

    Any luck finding that passage you mentioned (the one I can’t find in my copy, although p120 is part of the chapter on the Sun – it has me intrigued)?:

    OK, where are you up to in the book? Have you reached any of the pure bunkum science yet? Around page 120 he quotes the abstract of a paper that claims the sun is made up of the same composition as asteroids / meteorites.

  25. But John the BOM didn’t get it wrong, those who misinterpreted their statement did.

    40% chance of above average = 60% chance of below average
    60% chance of below average = 40% chance of above average

    Interchangeable and one and the same to me. The fault lay with the reporter as you point out and rightly attempted to give the correct interpretation. If they are too thick to see it’s one and the same, the fault lies with them not the BOM.

    Deliberately optimistic? I suppose if they report the other way they are deliberately pessimistic. And why would the BOM want to deliberately have their forecast construed as optimistic? What do they gain or achieve from doing that apart from having you contacting reporters to put them to rights?

  26. Good post Tom and I agree. ETS, especially in the terrible way the Rudd government is planning to implement it, is not the long term answer and can be actually worse that what it is purporting to cure.

    What is also bad is the Rudd has cornered the opposition into me-toosim on this so they are more or less proposing the same thing but with some parts even worse. So now we have no viable opposition to this nonsense or for any alternatives or compromises.

  27. Thanks MB.

    I think the entire debate has become unsatisfactory and political, and simply about ETS. The ETS is no panacea, the government uses it as a wedge, and the opposition get into the mud and scare campaign.

    Anyone would think that the Greens would be talking about sensible conservation measures and a public education/encouragement program, but they are just scoring anti development points.

    Pathetic.

  28. Mobius Ecko, on May 25th, 2009 at 11:06 am Said:

    “But John the BOM didn’t get it wrong, those who misinterpreted their statement did

    Deliberately optimistic? I suppose if they report the other way they are deliberately pessimistic. And why would the BOM want to deliberately have their forecast construed as optimistic”

    Mobius, I disagree the outlook was for BELOW average rains, so it should have been qualified as to % chances of such. It’s then neither pessimistic, nor optimistic, but factual.

    You probably know yourself, that sometimes you might try to break bad news to someone by being indirect or even evasive.
    As to why BOM might want their bad news presented in a more palatable way, that might be connected to the decision made by the Japanese IO researchers in not issuing a contradictory forecast in 2007 regarding BOM’s forecast of La-Nina. One of the main reasons the Japanese decided not to issue which was quite a gloomy outlook, was because they feared that it might push some Australian farmers, already under a lot of pressure, over the edge into committing suicide. Don’t take my word for it, it was reported in the “Weekly Times” when they got onto the story and probably can be verified by a search on the web. Get onto the “Weekly Times” if you like and ask for any issues contain articles on the Indian Ocean Dipole over the last year, 18 months, it’s all there.

  29. The latest weapon in the Denialist Armoury: Geoffrey Chaucer (from the Porlogue toThe Canterbury Tales):

    “Whan that aprill with his shoures soote
    The droghte of march hath perced to the roote,
    And bathed every veyne in swich licour
    Of which vertu engendred is the flour;
    Whan zephirus eek with his sweete breeth
    Inspired hath in every holt and heeth
    Tendre croppes, and the yonge sonne
    Hath in the ram his halve cours yronne,
    And smale foweles maken melodye…”

    Hah. See. No doughts. The sun shone. It rained and the crops grew in merrie Olde England.

    So Climate Change must be bollocks, right?

  30. So the BOM by sounding more optimistic was attempting to do the right thing in its mind because of rural depression and suicides and you contacted a reporter to ensure the more pessimistic sounding way of reporting the same data was given.

    Or have I missed something?

  31. @Tony:
    As mentioned, contact me offline if you want to be looking at page-by-page references. I think a total of two people on this forum have this book – so us tic-tacking back & forth at that level is a distraction to the larger debate.

    What you are looking for is the details regarding the sun as a collapsed supernova, which is almost word-for-word the abstract of this paper. Don’t know where I can find a non-pay gateway for the article, but I’ll send it over if/when I find it.

    I happen to know from friends of mine (academics) that the paper quoted is bunkum and is widely ridiculed in astronomy circles. Of course, with this information – you can ignore me (and my untrustworthy opinions) and look it up yourself 🙂

  32. I happen to know from friends of mine (academics) that the paper quoted is bunkum and is widely ridiculed in astronomy circles.

    Interesting. Do you happen to know if this paper was “peer-reviewed”?

    (Thanks to your clarification I can confirm the O. Manuel , S. A. Kamat and M. Mozina paper was cited on p120, footnote 516)

  33. I wouldn’t know 100% (not being the astronomy geek), but I’ll send her an email today and get back to you. Nag me if I forget (blogging is something around 10th priority, down the list from family time & a little up from testing code 😛 ).

  34. …testing code…???

    What kind of freak programmer are you, Ben? You keep this up and you could give us programmers a decent reputation.

    If my code compiles cleanly then I assume that it is OK – after all, I just treat production as a test bed for my programs. Far better to get end users to waste their time testing my programs – 😉

  35. Well, there we go. A handful of supposedly lateral thinking exemplars that amount to little more than circular inductions, johnd almost admitting that on logic alone his position is untenable, and Sparta begging the question again on his golden palamino hobby-horse du jour. Me, I’m going to run the video of a decompositional NEO dropped into the orbital path to both expand the NASA budget and to dilute the atmospheric CO2, just for fun; maybe call it Armageddon 2: Beyond The Limits of Delimited Logics.

  36. B.Tolputt
    I haven’t been able to find the solar info in any freely available source but the ridiculing of the premise seems entirely logical and it is not unreasonable to take the scientific expertise that the premise is scientific nonsense and proven so.

    Also the whole solar radiation theory of it being cause for the warming in the last two hundred years has been well and truly refuted, with some of the scientists who first proposed it recanting or going silent on the subject after the refutations were published.

    One exception is a Swedish study for which I posted a link a long time back. I will see if I can resurrect it as it also put paid to this funding furphy.

    Basically they have come up with a viable theory on solar radiation amplification through cloud forcing and have been given considerable government funding to research this phenomenon, including building a cloud chamber to model atmospherics and radiation input. This might prove why the solar radiation models on their own always fail without some type of atmospheric amplification. But even so the scientists have admitted their theory will not account for all of the heating but it might be an answer to the unprecedented rate of warming as there is now increased cloud cover due to man’s activities, so it still might come back to being anthropogenic.

  37. Mobius Ecko, on May 25th, 2009 at 11:50 am Said:
    “Or have I missed something?”

    Mobious,
    Somebody has to take the hard and lonely road to ensure people are told what they need to hear, and not what they want to hear.
    I know, I know, you don’t have to say it, thats not what true believers want, a dose of cold air upsets their warm and fuzzy feeling.

  38. Woohoo! Now we have a hook for me to put up the clip on Solar (which I will do this evening).

  39. One day someone will answer the simple question of why the temperature graph has headed east whilst the CO2 graph continues north. And one day someone will answer the question of how long before a continued eastward trend on the temperature graph, in the face of a continued northern trend on the CO2 graph, can be held as evidence that CO2 is not the main temperature driver such that “man made CO2 is definitely mostly responsible for the warming of the last century”. It is not an answer to say that 10 out of the last 12 years are the hottest on record, that simply says that at most, temperatures have plateaued at a recent high, and is not evidence that they are continuing to skyrocket in line with skyrocketing CO2. It’s reached a point now where there’s little value in quoting Bolt or Lambert, or even delving into their sources, because each side will say that the other has been discredited and the likes of us cannot possibly know. Karoly’s “10 out of 12” is evidence of climate stability, not climate change, so he’s no good. I have heard these cries of “wolf” so many times in my lifetime that I need something that I can see with my own eyes. Tell me how long the current stability will last before spiralling off into sharp warming. It’s been nearly a decade now (more than if you include 1998). Will the temperature start climbing again next year, the year after, or in 10 years?

  40. Mobius Ecko, on May 25th, 2009 at 1:22 pm Said:
    “there is now increased cloud cover”

    Mobious, you don’t think they might be onto something?

    Ah, increased cloud cover,
    warmer nights,
    no frosts,
    cooler days,
    more rain.
    Perhaps even global cooling?
    Are you sure there is increased cloud cover? World wide?

  41. No John wrong. Please read up for yourself or do you already know and are fishing again?

    Anyway there is nothing wrong with weather around the world, nothing at all:

    http://www.crikey.com.au/2009/05/25/the-worlds-meteorologists-report-from-the-frontline-of-climate-change/

  42. James
    I posted a link from a German climatology centre that explains it. It has scientific sources you can read up on.

  43. johnd, on May 25th, 2009 at 1:23 pm

    Ah now the arrogance comes to the fore, I was wondering when it would with you.

  44. Adrian. Where?

  45. James of North Melbourne, on May 25th, 2009 at 1:44 pm Said:
    “Tell me how long the current stability will last before spiralling off into sharp warming”

    The greenhouse effect relies basically of two very precise temperatures. One is the temperature at which water turns to vapour, and the other is at which it condenses. Each molecule absorbs a precise amount of heat at a precise temperature, transports it aloft, and dissipates it at a precise temperature. If nature had set out to intentionally build a temperature stabilising system for a greenhouse, of all the tools available to her she could not have picked a better one than water vapour. Each water vapour molecule is 400% more efficient at transporting heat than say a CO2 molecule plus there is just SO much of it just hanging around.
    The way I see it then as to whether the temperature can be stabilised or not comes down to volume of water vapour circulating in the system. I guess nature has worked out a way of making sure the right amount is in circulation at anyone time, probably does what we all do, puts what we don’t need immediately on ice. At times the water vapour probably has to work hard, rushing to absorb heat at the surface, rushing upstairs to let the heat go, then straight back down ago to start the process all over again. It can’t afford to hang around up there, not like that lazy CO2 which has stuffed itself full of heat, and then just sits there, for YEARS, depriving all the plants down below of a critical building block, plants that are patiently waiting for the CO2 available to them to double in order for them to reach their optimal growth rates. But water vapour really looks after them, bringing moisture, keeping the temperature reasonably stable, shading them from the sun.

    James, sorry to ramble on, but with regards to your question, I believe that whilst the water on this planet remains constant, the temperature will remain within certain parameters. At times the weather events may become violent as the heat needed to be transported taxes the capacity of the available water vapour and it has to be moved in a hurry. But whilst the earth still rotates, and tilts on it’s axis, the system still has time to slow down and rest giving it some reserve capacity. I think any runaway temperatures will only come after the planets water reserves are depleted, so I don’t think it will ever worry too many of us.

  46. @James of North Melbourne:
    Adrian is posting as “Mobius Ecko”. No, it’s not a sock puppet – he acknowledged it very early on.

    @Solar Inquisitors:
    I think I have found a source for the paper that is not behind a paid gateway. I am always looking up research papers for my job (you’d be amazed at what I have to code up sometimes!) and it is sometimes easier than you would think to find a free copy of university research. I’m just checking that the copy I have found is indeed the paper being talked about in Heaven & Earth.

    I doubt any of us here could make a head or tail of it though – it assumes alot of astronomy & math knowledge. Even with my electrical engineering background (some heavy math once you get beyond the basics) – some of it is still gobbledy-gook to me.

  47. Mobius Ecko, on May 25th, 2009 at 2:23 pm Said:

    Come on Mobius, don’t be so thin skinned. Enjoy the cut and thrust.

  48. I think any runaway temperatures will only come after the planets water reserves are depleted, so I don’t think it will ever worry too many of us.

    Um, you have heard of the Ice Ages yes? Unless I’m missing some important piece of geological / astronomical knowledge where we somehow lost & gained vast quantities of water before & after said Ice Ages, even thirty seconds thought blows that concept clear out of the water (no pun intended!).

    This is why we need to trust the scientific process rather than popular opinion. Anyone can come up with such garbage, but some people are willing to ignore logic or simply don’t have enough facts to rebut this.

  49. @joni:

    If my code compiles cleanly then I assume that it is OK – after all, I just treat production as a test bed for my programs. Far better to get end users to waste their time testing my programs

    You work for Microsoft don’t you?!

  50. @Solar Inquistors:
    OK the PDF of the paper in question is here. I have been informed that it isn’t the exact paper that is published, as there were some minor changes since the paper was made available for commenting, but that it is probably as close as I’m going to get online without having a subscription to the research journals in question.

  51. Mobius Ecko, on May 25th, 2009 at 2:20 pm Said:

    “Anyway there is nothing wrong with weather around the world, nothing at all:”

    Mobius, a bit of sensationalist reporting, but really that is just the weather doing it’s job, isn’t it.

    As the worlds populations grows obviously more people are going to be affected by adverse weather events. People generally also tend to crowd onto the most fertile land which most likely gained it’s fertilty by the same sort of weather events that are now trying to drive people off.

    It’s a similar story to the farmer who tries to run 10 sheep to the acre and who is almost always in drought conditions, but his next door neighbour who only runs 4 sheep to the acre is never in drought conditions.

  52. I have never been so offended in my life!

    (yes yes yes reb – I know – I must get out more often)

  53. B.Tolputt, on May 25th, 2009 at 3:29 pm Said:
    “Unless I’m missing some important piece of geological / astronomical knowledge where we somehow lost & gained vast quantities of water before & after said Ice Ages,”

    Ice is one of the three forms of water as far as I’m aware.
    Unless water travels to and from outer space I believe the planets H2O inventory remains constant. Correct me if I am wrong.

  54. John,

    I think it is this line that causes the problem:

    I believe that whilst the water on this planet remains constant, the temperature will remain within certain parameters

    Unless you are saying that Ice Ages are within the certain parameters.

  55. Thanks joni. That is exactly what I was getting at and I get the feeling johnd knew that. Of course, I’ve been known to be wrong occasionally… whenever it happens though – I get told it is a sign of the Apocalypse 😛

  56. I am never wrong, it is just that others ask the wrong question. That is why I am known around work as FIGJAM. hehe

  57. James sorry to not get back to you on the German Climatology centre link but I can’t find it anymore. Does anyone know which thread I posted that link (if I posted it in this blog, which I thought I did)?

    Anyway I found this lot of heavy reading to go on (and I’ve only skipped over some of it) out of lots on the subject:

    http://climatechange.flinders.edu.au/NS%20CO2&warming.doc
    http://data.giss.nasa.gov/gistemp/2007/
    http://www.cru.uea.ac.uk/cru/data/temperature/

    In the meantime I’ll keep looking for where I put that link.

  58. Adrian, given this….”Furthermore, recent sighting of the first sunspot of reversed polarity (reported Jan. 4 by, e.g., SpaceWeather.com and NOAA) signifies that the ~ 4-year period of increasing solar irradiance is about to get underway.”

    …from the second of your links, for the purposes of the debate, can we settle on 2012 as a date where AGW is proven or otherwise. Stuff all will happen politically in the meantime as we all know, but if temps don’t return to their pre 1998 (or 2002) northward trend, can I then confidently say it’s a crock?

  59. My wife bought be FIGJAM t-shirt once… when we were still teenagers. Becoming a code-monkey has, obviously, made this aspect of my personality much more subdued… *cough*hack*splutter* Sorry, choked on my own sarcasm there 😛

    Kamahl: Back on topic please folks…

    B.Tolputt: Kamahl, why are you so unkind?

  60. B.Tolputt, on May 25th, 2009 at 3:40 pm

    Good work finding that paper. Now I wonder if you could trouble your friend further for help finding the reviewers’ comments. I am extremely interested to see what the famous peer-review process thought at the time of a paper now considered by the illuminati to be “bunkum science”.

  61. Doubt that I’d get a hold of them, Tony. Most “peer-reviews” for scientific papers do not publicly publish the reviewers comments, simply sending them to the author &/or the editor of the journal in which they are trying to be published in.

    There are only a handful of scientific journals that publish the peer review comments, and maybe one or two worldwide that publish the names of the reviewers (i.e. “open peer review”).

    I would get a layman’s review of why the paper is bad, but I’d rather not expose my friends to the ridicule I expect given the audience is using terms like “illuminati”, “leftoids”, etc. Given that kind of language, and the continual use of the “conspiracy of scientists” that pops up like a bad penny every ten-to-fifteen posts – this is not a forum debating science; it’s politics as usual with science as the football.

  62. joni, on May 25th, 2009 at 3:56 pm Said:
    “I think it is this line that causes the problem:”

    joni,
    I can see your point.

    Yes, ice ages are within the parameters, how can they not be? As long as the planet had an atmosphere, H2O and temperature differentials that spanned the points at which a change of state occurs.
    As far as we know, the entire planet was never completely covered in ice during the ice ages, thus there had to be an uneven distribution of heat between regions and altitudes.

  63. Yes, ice ages are within the parameters, how can they not be?

    Well, given those parameters, I think it would be possible that “runaway temperatures” as most humans would define them are possible and hence worrisome.

  64. B. Tolputt 6:23pm,

    So where does that leave the peer-review process?

    If, as you say, peer-review is used when “they are trying to be published”, this paper has evidently survived that process: it was published in Astrophysics 654: 650-654 (according to Plimer’s footnote).

    So the question is: If such “bunkum science” isn’t culled by the process, is peer-review really the infallible filter of scientific thought it is purported to be?

  65. Actually, from what little I can gather – the journal in which it was published (which was actually “Physics of Atomic Nuclei”, not “Astrophysics”) is a Russian-based journal with lax (or non-existent, depending on who you talk to) review standards.

    Like all areas of the media, scientific journals have good publications and dodgy ones. For example, whenever I hear someone read it “in the paper”, I ask “Which one?”. We all know that some newspapers are more reliable &/or less biased than others – even with the requirement to back up their publications with enough fact to prevent lawsuits. Same goes for science journals.

    In this case, we have a US academics reaching overseas to find a journal willing to publish their work. From what I have heard & read (which is admittedly little given I’m not an astrophysicist), he somewhat embarrassed his university with the publication.

    I have been pointed to an online source of debate with an author of paper, but most of the details being argued are so far out of my base of knowledge, I can’t know who is making better sense…

    Debate being held here. Be warned though – the people discussing the subject are more knowledgable than I’d say any of us are on the topic.

  66. Tony, peer review also means that people publish both comments on papers and papers addressing – even refuting – previously published papers.

  67. B. Tolputt,

    Actually, from what little I can gather – the journal in which it was published (which was actually “Physics of Atomic Nuclei”, not “Astrophysics”) is a Russian-based journal with lax (or non-existent, depending on who you talk to) review standards.

    Not according to Plimer. Maybe that’s another point you can add to his litany of errors?

    Debate being held here. Be warned though – the people discussing the subject are more knowledgable than I’d say any of us are on the topic.

    One thing stands out about above all else at that forum discussion: There is certainly no consensus on the topic – in fact just the opposite (fierce disagreement would be a better description).

    By the way, as a matter of fact only, the debate is not “being held” – all 13 pages are dated 2006.

  68. B. Tolputt, Lotharsson,

    Michael Mozina (co-author):

    You should respect my effots because these are peer reviewed publications, that go out to professionals with the expertise to judge the validity of our work. These papers were not simply submitted to generic confernces on topics as was the case with the first paper I was involved in. There is a process here Dave that every scientists uses to get their matieral published and to make it widely available and to open in to significant scrutiny. That’s all we’ve done.

  69. Well that didn’t work. Try this.

  70. Read further, Tony. People have asked for proof of the peer-review in which it was published. The requests have been ignored. Try as I might, I cannot find the guidelines either.

    This is not to say they are not there, but the issue is/was in contention. It’s a bit late for me to be bugging my friends on this right now too.

  71. Yes, very interesting, and, whether the theory is a valid one, or “bunkum” as you suggest, I am fascinated by the peer-review implications.

    A couple of things come to mind:

    In which journals does a paper need to be published to legitimately claim peer-reviewed status, and who determines this?

    Do “bad” papers slip through the cracks, even in the ‘legitimate’ journals? In other words, how stringent is the process?

    Do all reviewers tackle their work with the same rigour? (Obviously not.)

    Do reviewers recieve a fee?

  72. Also, in narrow fields of research, where all participants are known to one another, is it possible to have impartial reviews?

  73. Incidentally,

    Another, less ‘aggressive’, discussion on this paper here.

  74. Tony,

    Yes, bad papers sometimes slip through the cracks, sometimes even in good journals. The wheels of science may grind slow, but they grind exceedingly fine – over time most bad papers get detected and refuted. A peer-reviewed paper is a necessary condition to be taken seriously, but not sufficient to say that it’s conclusions are valid. For one thing, a viable refutation of (say) a hypothesis held in one paper may not be immediately obvious, or it may take quite some time to do the corresponding work…

    I’m sure all reviewers don’t tackle their work with the same rigour.

    No, peer reviewers do not get paid for it (at least I wasn’t the case when I did it).

  75. Do “bad” papers slip through the cracks, even in the ‘legitimate’ journals? In other words, how stringent is the process?”

    Yep i remember those Korean scientists who got caught

    http://www.redorbit.com/news/science/339702/south_korea_tries_to_pick_up_pieces_from_science_fraud/

    They got caught because someone tried to reproduce the work and couldn’t. The work was published in Science.

    Interestingly something similar happened with the Nature paper with the hockey stick graph. The Canadians didn’t agree with the findings of the hockey stick paper.

  76. Thanks for that link, Tony. Seems that there is a history of aggression with this guy. The thread I linked to was indeed more aggressive, but had alot more detail as well (hence my warning).

    The sample of three threads doesn’t bode well for the objectivity of the Micahel Mozina (or the acceptance of his paper / science); I’ll give it that much.

  77. “Ice is one of the three forms of water as far as I’m aware. Unless water travels to and from outer space I believe the planets H2O inventory remains constant. Correct me if I am wrong.”

    Spot on……but logic is wasted here……..I am suprised nobody has asked for a model or link proving it…..LOL…

  78. Spot on……but logic is wasted here

    Yup. No matter how much logic is applied & demonstrated, Sparta just doesn’t get it.

    It’s just not worth the effort, after all – he completely misses the point that an Ice Age might be a worrisome. Reading further than something he can make a smart-ass remark on is just beyond him.

  79. B.Tolputt, on May 25th, 2009 at 10:36 pm Said:
    “Ice Age might be a worrisome.”

    Ben, yes, extremely worriesome. To lose all those degrees gained coming out of the last one, what, 10 degrees more or less. You’d wonder why anyone would get concerned about a fraction of a degree further rise.

  80. Is anyone knowledgeable on Greenland?
    I’ve read some stuff about how temperatures were higher there in the 1920/30s, starting falling until about a decade ago, and then began the current rise bringing them back to the temperatures last seen there 90 odd years ago. Who are the reliable sources on Greenland’s climate?

  81. Haven’t been following this closely as the early response from scientists of Plimer’s book set off my bulls**t detectors, but I noted this:

    Ice is one of the three forms of water as far as I’m aware. Unless water travels to and from outer space I believe the planets H2O inventory remains constant.

    Not sure why it matters whether it’s constant or not (haven’t been following closely…)

    But firstly water can be disassociated into Hydrogen and Oxygen, which can then combine with other atoms to form molecules which are not H2O. And combustion of certain compounds (alcohols? Chemistry was so long ago) produces water as a side product.

    So, no, the H2O inventory doesn’t necessarily have to be constant. Maybe these types of processes change H2O levels insignificantly on a global scale, maybe not – I don’t know.

  82. Also, in narrow fields of research, where all participants are known to one another, is it possible to have impartial reviews?

    Science is ultimately adversarial by design (in principle on the level of hypotheses, but in practice people tend to get a bit personal about their pet theories at times).

    So in a small field, and even in large ones, impartial reviews are not necessarily possible. But then pre-publication peer review is meant to assess whether it meets minimum standards for publication, not whether it’s 100% correct. Reviewers must provide their reasoning in their feedback, and thus even a non-impartial reviewer can provide good enough feedback for a competent editor/publisher to make that decision fairly well (most of the time).

  83. The number of “scientists” somehow adds credence to the argument; all with paychecks now dependent on that “support”? I am sure it has nothing to do with grant funding or anything, it is all in the interest of mankind……

    Is someone really still trotting out the old “grant funding biases scientists towards supporting AGW when the evidence isn’t there” canard? I didn’t see the original context, so if I’m wrong I apologize.

    But assuming my interpretation is correct, think about it.

    How many vested interests would LOVE to prove that there is no AGW? (Heck, I would. It would make life a lot simpler in many respects.) And how cashed up are some of those interests? If you’re a scientist and you have a plausible idea that you reckon you can do some work on and robustly disprove AGW, you can get as much money as you need to do the work, no questions asked. (Not to mention the Nobel prize that is likely waiting for you…)

  84. Lotharsson,

    “So, no, the H2O inventory doesn’t necessarily have to be constant. Maybe these types of processes change H2O levels insignificantly on a global scale, maybe not – I don’t know.”

    Now you’re a chemistry major, hilarious. Water is a “compound” and as the comment was about the “compound” water, not its elemental constituents your statement reflects you usual MO. A compound that exists in 3 forms (gas, liquid and solid) as mentioned (Earth being especially unique as being able to support all three forms simultaneously). If you change H2O’s composition in any form (not talking about protons) it ceases to be water…..Understand?……(take a look at the “water-cycle”).

  85. “Is someone really still trotting out the old “grant funding biases scientists towards supporting AGW when the evidence isn’t there” canard?”

    Clearly somebody has never worked in academia? As usual you miss the point. You don’t have to necessarily prove one way or the other and the point in academia is usually to do neither in most regards. Once a “theory” gains some traction most grant funding is based upon studies that measure outcomes on a whole host of possible “repercussions or gains”. In most cases, as long as a particular study “purports” to look at variables that may have some bearing one way or another, the funding becomes a lot easier to attain……This is hardly a novel concept but I am sure you knew that and as always, you are just looking to argue about something…LOL…….

  86. B. Tolputt,

    Mate, your picking a fight with the wrong person….I simply “agreed” with a very logical statement. Don’t get upset with me, jus think before you type…..

    Johnd,

    “You’d wonder why anyone would get concerned about a fraction of a degree further rise”

    Absolutely golden…………

  87. Lotharsson, on May 25th, 2009 at 11:51 pm Said:
    “Not sure why it matters whether it’s constant or not (haven’t been following closely…)”

    I feel that it is important. Each of it’s forms stores a different amount of heat, so as it changes state it either releases or takes up heat into or from the environment. It also transports heat also in each of its forms, ocean currents, evaporation, even icebergs and sea ice, as it seeks to redistribute the heat caused by uneven heating of the planet. A certain volume can only do a certain amount of work. At one extreme if it was to give up all it’s heat, all water would turn to ice, once it reaches absolute zero it’s exhausted. At the other extreme it can only continue to absorb heat until it all turns to vapour. Fortunately whilst in a form other than liquid, it has properties that regulates the primary source of heating, solar energy, but it’s capacity to do so is finite.
    If the planet was to lose water then that overall work capacity would diminish.

  88. Sparta of Phoenix, AZ USA, on May 26th, 2009 at 2:07 am Said:

    We both know it’s all about perspective, and keeping our feet on the ground. Cheers. 🙂

  89. Has anyone got the accompanying notes for this graph. I am looking for the explanation for the yellow sector. I assume, given it is for the period where temperatures have been reconstructed, it represents the range of the data they have for that period.

    Also is anyone continuing to publish proxy data for the period covered by instrumental data. It would be interesting to see what modern day results would be thrown up using such limited and reconstructed data, as well as providing ongoing validation of assumptions regarding such data.

  90. Did any of those discussing climate change watch Jamie Durie on Sunday Night and his visit to the Inuit people of Canada and thier elders.

    If not I suggest you watch it. I would much rather believe the elders of a nation that has existed on the ice for centuries, with their history handed down for generations than hypothesies from all different angles.

    These people are living the effects.

  91. These people are living the effects.”

    Shane- any evidence that this effect is due to humans??? The theme of Plimers book is that climate is cyclical. This would have happened anyway according to Plimer.

  92. Shane,

    Did those Inuit elders have any opinion on whether the changes they are witnessing have anything to do with human-induced CO²? If so, what part of their handed-down knowledge did they draw on to arrive at those opinions?

  93. Now you’re a chemistry major, hilarious.

    Now you’re lying again. Not hilarious.

    Water is a “compound” and as the comment was about the “compound” water, not its elemental constituents your statement reflects you usual MO. A compound that exists in 3 forms (gas, liquid and solid) as mentioned (Earth being especially unique as being able to support all three forms simultaneously). If you change H2O’s composition in any form (not talking about protons) it ceases to be water…..Understand?……(take a look at the “water-cycle”).

    Yes, dude, I already understood that. It’s you that didn’t get the point.

    The statement I was addressing was, and I quote:

    Unless water travels to and from outer space I believe the planets H2O inventory remains constant.

    (My emphasis.)

    In your rush to condemn me and what you think is “my usual MO”, your explanation supported my point.

  94. “In your rush to condemn me and what you think is “my usual MO”, your explanation supported my point.”

    That is funny…….

  95. Clearly somebody has never worked in academia?

    Are you saying you have? Or is this another “As a lawyer…” moment? Because I have a postgrad research degree, which means I’ve spent some time working in some fashion in acadaemia.

    And as is common you haven’t argued my point, just something you fancy I might have said.

    BTW, I’m assuming the original comment was about AGW. It’s not primarily that the _number_ of scientists that adds credence, although that definitely has a credence-adding effect. (Science is, after all, adversarial, and huge careers are made by disproving the commonly accepted explanations.) It’s the range of their disciplines and work that all in different ways and via different paths converge to the same broad set of conclusions.

  96. At the other extreme it can only continue to absorb heat until it all turns to vapour.

    Check into this. Unless I’m very much mistaken, steam can get hotter than boiling point – much hotter.

  97. That is funny

    I thought so. Glad you could share the joke.

  98. Lotharsson, on May 26th, 2009 at 10:22 am Said:
    “Check into this. Unless I’m very much mistaken, steam can get hotter than boiling point – much hotter.”

    In the atmosphere? Really? Does IPCC allow for this? You might be onto something significant. 😉

  99. Neil

    Did you watch the report ?

    Tony

    Watch the report and judge for yourself.

    It is fine for you both to be sceptics but rather than ask me ( for which you will then agree or disagree based on your own opinions), watch the report and make up your own minds.

  100. I would assume they understood the principles, johnd 😉

  101. Did those Inuit elders have any opinion on whether the changes they are witnessing have anything to do with human-induced CO²?

    In one important sense, I’m not sure that really matters.

    A major part of the debate over climate change that often gets left behind due to the debate over how much humans contributed to it, if at all.

    Let’s assume for the sake of argument that essentially all currently observed climate change is due to “natural causes” and no human influence contributed in any significant way.

    Fine. So given that assumption, what do we do about the upcoming climate change? It’s going to potentially cause major disruption to many many people. There are an awful lot of people living just above sea level. Climatic zones will likely migrate away from the equator (e.g. the US wheat belt may migrate to Canada, but the soil there is said to be much poorer and so obtaining the same yields as the US does is likely to be extremely difficult). And so on…

  102. Lotharsson, on May 26th, 2009 at 10:48 am Said:
    “Fine. So given that assumption, what do we do about the upcoming climate change?”
    Firstly, working out the direction of the change would come in handy.

  103. So given that assumption, what do we do about the upcoming climate change?

    Here’s an idea: we could spend less time and money on trying to “stop” catastrophic natural climate change, and more on contingency planning.

  104. …we could spend less time and money on trying to “stop” catastrophic natural climate change, and more on contingency planning.

    Is it your assumption that natural climate change is by definition not catastrophic? For which species or ecosystems do you think that holds?

  105. Firstly, working out the direction of the change would come in handy.

    Do you think there’s any serious doubt about the direction?

  106. Is it your assumption that natural climate change is by definition not catastrophic? For which species or ecosystems do you think that holds?

    Whether it is catastrophic or not remains to be seen (my instincts say not).

    The kind of hyperbole being used by AGW proponents – “dangerous”; catastrophic” – is nothing more than alarmist fearmongering doomsaying, designed to regain the attention of an increasingly uninterested public.

  107. That is funny

    I admit I’m not entirely sure if it’s the real Sparta (whomever that may be) or a satirist at work, but I’m not sure what was funniest.

    Perhaps it was that – in response to my post pointing out (admittedly implicitly) that water is a compound which can be both created and destroyed (by chemical reactions such as “disassociation” and “combustion”) – that you felt the need for a condescending little lecture on the fact that water is a compound.

    Or maybe it was the fact that you responded to my comment pointing out that the total amount of water in the world need not be constant because of those chemical reactions that create or destroy water – by pointing out that “changing the chemical composition” of water would mean it’s no longer water.

    Or maybe it was that you referred at the end to “the water cycle” which has nothing to do with chemical reactions that create or destroy water.

    It’s too hard to decide.

  108. “And as is common you haven’t argued my point, just something you fancy I might have said.”

    If there is anything I have gained from our past “engagements” is that it is best to just let you chase your own tail….Often, you have no point or one that you have difficulty keeping track of; one that needs consistent reiteration. Really, as you once said to me, what is the point…But just to give you an idea, please do a Google search on “Global Warming studies” and effects on insects, plants, etcetera….it is practically endless. No, I am just crazy for purporting such a thing indeed……trotting out more reality I guess, my apologies….Here is just one of many….the effects of concrete and AGW no less, mighty useful I am sure….

    http://www.azonano.com/news.asp?newsID=3626

  109. The kind of hyperbole being used by AGW proponents – “dangerous”; catastrophic” – is nothing more than alarmist fearmongering doomsaying, designed to regain the attention of an increasingly uninterested public.

    Depends on your definition of “dangerous” or “catastrophe”, I guess, and whether your instincts are right or the scientists are.

    I’m not sure the public is “increasingly uninterested”. I tend to think that AGW has pretty much mainstream acceptance these days.

  110. But just to give you an idea, please do a Google search on “Global Warming studies” and effects on insects, plants, etcetera…

    Sparta, please stop it before I die laughing.

    In response to a minor point about the total amount of H2O in the world…you end up telling me to search Google – which you’ve previously disparaged when I’ve suggested you use it – for global warming studies of effects on insects, plants etc? Pray tell, how do these studies of insects and plants relate to the amount of H2O on the planet, and whether it is constant or not?

  111. Sparta, sorry, still had my head in the H2O space. That last comment of yours referred to a different thread of discussion.

  112. “Is someone really still trotting out the old “grant funding biases scientists towards supporting AGW when the evidence isn’t there” canard?

    Again, why it is fruitless to engage with you, you simply can’t keep track of your own comments…..LOL…NO worries, I have also worked with the schizophrenic….I understand completely….

    “Or maybe it was that you referred at the end to “the water cycle” which has nothing to do with chemical reactions that create or destroy water.”

    Sigh, take a deep breath and go back and review what you have posted recently before forging ahead….

  113. “Sparta, sorry, still had my head in the H2O space. That last comment of yours referred to a different thread of discussion.”

    No offense mate, but you do have trouble keeping track of what were discussing most of the time. Again, it makes more sense to debate one topic at a time….You sure your not related to Adrian?

  114. Sparta, your “clearly you haven’t worked in acadaemia” comment was in response to me saying you hadn’t argued my point. “My point” referred to laying out an interpretation of an earlier statement (probably made by you on a different thread), and arguing against it.

    That interpretation of the earlier statement was the idea that somehow science funding biases researchers to produce results that support the AGW hypothesis.

    In response, you didn’t argue my point, which – to clarify – was that:
    (a) There’s clearly a market out there for research results that disprove AGW, so anyone who wants to do so and has a plausible idea could certainly find oodles of funding. (In fact, it may be easier to do so than to find funding for “routine” research on the edges of AGW…)
    (b) Scientists are a notoriously competitive bunch who love to take down someone else’s theory, so they have plenty of motivation to do so.
    Therefore it’s hard to see that funding mechanisms bias results on the question of whether and how much humans have contributed to global warming.

    In response you cite via Google what I assume are studies of the potential effect of any global warming on insects and plants etc. I assume you do this to point out that plenty of scientists are not working on whether AGW stands up to scrutiny or not, but on what the effects will be.

    Well, of course! But those scientists do not get counted in the tally when people are determining how many scientists’ work supports the AGW hypothesis, precisely because they’re [b]not addressing that question[/b]! This is further illustrated by the fact that they don’t NEED AGW to be true for their work to be useful. We really need to understand the effects of climate change, whether “natural” or “anthropogenic” or a mix of both…)

    And if those scientists’ work are cited by someone as “supporting AGW” then I’ll be right behind you protesting that they do not.

  115. shaneinqld, on May 26th, 2009 at 10:42 am

    So the Inuits just know that this must be due to humans. They must be leftoids.

    But I havn’t watched the report. Just to play devils advocate. Did the lefty reporter (Durie) try to slate the story his way??

    I must admit my respect for the media is at a new low so i don’t tend to trust what people say any more.

  116. “somehow science funding biases researchers to produce results that support the AGW hypothesis.”

    No, I never said that they were looking to produce results that support AGW because of funding, I was simply implying that the amount of research being conducted presently in its name or related to it is certainly not to be had without said theory………Hence, they have more to gain from it perpetuation don’t they?

  117. “This is further illustrated by the fact that they don’t NEED AGW to be true for their work to be useful. We really need to understand the effects of climate change, whether “natural” or “anthropogenic” or a mix of both…)”

    Yes, but they NEED AGW to get funding is my point, they simply wouldn’t have “work” if not for the AGW debate, understand…..NO AGW debate, no funding…….Now we need to understand climate change “natural” or “AGW”? Oh yes, that movie by Al Gore, right…..Yes, well my weather man is right maybe 30% of the time…LOL…….I certainly don’t think we should be funding research based on his forecasts either….

  118. No, I never said that they were looking to produce results that support AGW because of funding,

    OK, fair enough. But you could have just said that in response to me laying out the interpretation and then saying:

    I didn’t see the original context, so if I’m wrong I apologize.

    But assuming my interpretation is correct…

    Moving on.

    I was simply implying that the amount of research being conducted presently in its name or related to it is certainly not to be had without said theory………Hence, they have more to gain from it perpetuation don’t they?

    Most climate scientists believe the earth is warming. Whether it is due to human influence or not, humanity still needs to understand what the effects of climate change will be. And hence all of those scientists researching the effects of climate change do NOT have more to gain from the perpetuation of AGW.

  119. Yes, well my weather man is right maybe 30% of the time…LOL…….I certainly don’t think we should be funding research based on his forecasts either….

    Me neither, because it’s a common misconception that “weather” is “climate”. Even though there’s a relation between them, they’re two different things.

  120. Lotharsson,

    C’mon mate, you know what I mean…….

  121. C’mon mate, you know what I mean…….

    No, I’m usually not sure I do. Which comment are you referring to?

  122. “Which comment are you referring to?”

    The weather man analogy…..

  123. OK, the weather man analogy.

    I disagree with the conclusions that you draw from your analogy. We can do a better job of predicting climate than we can of weather, just like we can do a better job of predicting average traffic levels over five years than how bad tomorrow’s peak hour may be.

    And when there’s a decent chance the climate will change in ways that will be challenging to respond to, then the research is not only justified but prudent.

    But I doubt we’ll agree on these points.

  124. Neil

    Watch the report instead of your usual banter about leftoids.

  125. Neil

    Here is the link hopefully.

    http://au.tv.yahoo.com/sunday-night/video/-/watch/13630011/

  126. “Believe nothing just because a so-called wise person said it. Believe nothing just because a belief is generally held. Believe nothing just because it is said in ancient books. Believe nothing just because it is said to be of divine origin. Believe nothing just because someone else believes it. Believe only what you yourself test and judge to be true.”~Buddha

  127. Fun with mental schema 101 (more for Tony’s benefit than anyone else’s, because he might observe that I’ve been toying with the meta-model for some time).

    Yes, it’s astounding, Sparta, how we all gain from an emergent understanding of the concrete nano-properties of a near-ubiquitous material product that has been used for millennia without understanding those properties of it and are able to investigate new efficiencies for it.

    The noosphere (and the complex economies), arguably, appears to move in mysterious ways; is there a tectological domain where the polytelic endeavours of the scientifico-nooietic stigmergy aren’t self-organising new niches in the Grand Red Queen’s race through their participations?

    I was going to mention that there are four states of matter, and a variety of polymorphs and structural arrangements within those, too. I promise that I won’t catastrophise about ice-nine, though, in making that minor observation. But, then again, I did also mention back a couple of threads that what matters may be not the matter itself in the larger scheme of things. 😉

  128. Tony

    Are you a buddhist ?

  129. Are you a buddhist ?

    No shane, I’m a bulldust (artist).

  130. Tony, on May 26th, 2009 at 2:28 pm

    Prior to the recent collapse of the art market world eonomy, you’d probably have made a small or large fortune in that medium; perhaps wait for the next bubble to peddle your (es)c(h)atological wares.

  131. eonomy economy (es)c(h)atological (e)sc(h)atological

  132. Can’t wait, Legion. This bear sh*t is so hard to work with.

  133. Shane

    Thanks for the link. I looked at the report about the Inuits. While they have said that their climate is changing there was no evidence that humans had anything to do with it.

    Thats what Plimers book is all about. Climate has always changed.

  134. Thats what Plimers book is all about. Climate has always changed.

    If that’s what Plimer’s book is all about, he could have saved himself a lot of trouble. Climate scientists acknowledge that “climate has always changed”.

    Two guys are in a car doing 95km/h. They note they are approaching a very large rock wall, and slam on the brakes. We know that cars have always been able to stop if they see an obstacle in time. So are they safe?

    Well, firstly you need to know a bit more. Are they on a surface with good friction? Are the brakes working properly? Is the accelerator jammed at the same time? Are they falling straight down at 95km/h having just driven over a cliff? In other words, is this a “normal” scenario, or is there some important difference (maybe not immediately visible to the casual observer) in the state of the entities involved that may lead to a non-typical outcome?

  135. Plimer’s contentions, arguably, appear strongly inductive when stripped of their (un?)scientific gloss; it’s what bugged me most about Rattie’s weak inductive contentions, too; that tendency to avoid critical introspection and to resort to a highly delimited search when adducing inductive probabilities, especially when the conclusions posed are in a uni-dimensional ‘the boy who (de)cried the boy who cried wolf’ format.

    A skeptic might remain skeptical about both their claims, and entertain myriad other thoughts besides; including thoughts have-able about a wolf’s and a jackal’s similar seemings when seen in silhouette, and where only the (non-) existence of a wolf is within contemplation. As polemics, they’re fine (as a wise man once said…as contributions to a public debate); as definitive touchstones for, or of, anything, they’re about as comprehensive as the story of the Five Blind Men from Indostan, imho.

    Even johnd’s vaguely Gaian narrative about the goldilocks harmonics of an oscillating spring ignores that Hooke’s law only holds in proportion to the (in)elastic bounds. Which reminds me, what happened to the hook and the promised walk on the dark-side of the Sun?

  136. Which reminds me, what happened to the hook and the promised walk on the dark-side of the Sun?

    Was just thinking the same thing, however, I fear that golden thread may be en route – along with a small boy in a blue hat.

  137. What i find strange about this climate change thing is that it is so political. Leftoids (whoever they are) tend to believe in man induced global warming. Rightoids (whoever they are) are generally not convinced. Yes it may be happening but is it cyclical or induced by humans.

    Must be due to a difference in the way the leftoid and rightoid brain works.

    Climate change being science I don’t think this should occur.

  138. Neil of Sydney, on May 26th, 2009 at 7:21 pm Said:
    “Climate change being science I don’t think this should occur.”

    Is it science? Politics is certainly involved, but if it is science then their theories, strictly speaking, should have been proved by controlled experiment before being accepted as science. Is that not the criteria normally demanded by scientists themselves? Without that does any work receive their recognition?
    So far it’s all opinions and best guesses.
    The onus is not on those sceptical of the assertion of AGW. No amount of experimentation can prove the theory wrong, but one alone could prove it right, and make it science. (with apolgies to Albert)

  139. For those of us who dare to “question”, the horror….

  140. Climate change being science I don’t think this should occur.
    Riiight… because when religion (be it political or spiritual) clashes with science, science always wins out… That’s why we no longer have right-wingers trying to force creationism into science education programs. That’s why we no longer have people claiming they can “cure” the mental illness homosexuality.

    What’s that? We DO still have people doing those things? Oh well crap, seems that people don’t accept science that contradicts their personal viewpoint &/or the viewpoint of authorities they respect.

  141. @Sparta:
    Because, American op-ed news pieces are always so well researched… Like that CNN program with the chronologically challenged letter about G.W.Bush’s Reserve Guard fiasco or FOX News on just about every topic, every day.

  142. The onus is not on those sceptical of the assertion of AGW. No amount of experimentation can prove the theory wrong, but one alone could prove it right, and make it science.

    Isn’t that completely @rse about? I’m certainly no scientist, but was under the impression that a theory can never be proved true, only false. Isn’t that the basis of the scientific method? Karl Popper springs to mind…
    http://www.xenodochy.org/article/popper.html

    Karl Popper’s philosophy of science uses modus tolens as the central method of disconfirming, or falsifying, scientific hypotheses. Scientists start with a current scientific theory and use the usual methods of deductive reasoning to derive specific conclusions, of which some are “predictions”. Strictly deductive reasoning is “truth preserving”, that is, it is such that if one starts out with “true” premises, one can only deduce “true” conclusions. Starting with a “theory” and deducing “predictions” can be stated in the form of a premise:

    If the theory is true, then the prediction is true.

    Popper shows that we cannot prove that a theory is true, but we can certainly show that a prediction is false. If the scientist tests one of these predictions and finds out that it is not true, he uses good ‘ole modus tolens to conclude that the theory cannot be true.

    Or as Albert put it more succinctly:

    No amount of experimentation can ever prove me right; a single experiment can prove me wrong.

    http://www.brainyquote.com/quotes/quotes/a/alberteins100017.html

  143. bacchus, on May 27th, 2009 at 9:17 am Said:

    That’s Albert the sceptic speaking.
    Now what experiment could prove AGW exists? The assertion is that AGW is altering the natural climate cycle, not the opposite.

  144. …should have been proved by controlled experiment before being accepted as science.

    Bacchus is correct. You’ve got the idea of “science” arse-about. Science suggests hypotheses (often competing) for what is observed. (And what is observed is not restricted to “controlled experiment” – Popperian science is quite a limited subset of science itself.) Experiments and/or still further observations are used to test hypotheses. Those that survive are considered plausible; those that do not are discarded. It’s like a Darwinian competition of explanations – survival of the fittest (but don’t push the analogy too far – no reproduction here).

    This process is made more difficult by the inevitable noise in any measurement we make, which makes for interesting and extended debates amongst scientists because sometimes the data with its given level of noise isn’t clear enough to robustly disprove a certain hypothesis, and sometimes the data gathering methods introduces errors that aren’t fully understood and corrected for at the time.

    Those hypotheses that survive the most intense and sustained scrutiny tend to informally reach the status of “proved” (especially when no plausible competing hypothesis exists) – but this word “proved” is a piece of over-simplified (and convenient) popular slang that doesn’t correctly describe the status. Any single experiment or observation could prove even the most widely accepted (“proved”) hypothesis wrong.

    So far it’s all opinions and best guesses.

    That’s a rather … odd view of what it is so far.

    So far it seems to be the best explanation for the enormous masses of evidence we have gathered. The alternative explanations do not appear very likely to be consistent with observation, within the bounds of the noise in the measurements etc. That doesn’t mean AGW could not be proved wrong (as you say, perhaps even by a single well-chosed experiment or observation)…

  145. John,

    I’d have to agree with Bacchus and Lotharsson on this one, at least their point that the scientific method – Popper’s interpretation, anyway – is based on falsifiability.

    AGW, in my opinion, is falsifiable. If average global temperatures continually refuse to rise as predicted, while CO2 continues to rise, then the hypothesis is false. Of course Lotharsson touches on the wiggle-room that AGW proponents have worked into their theory: he calls it “noise”.

  146. For those of us who dare to “question”, the horror….

    There’s no problem with questioning, and therefore it doesn’t take any “daring”. Questioning is an integral and necessary part of science.

    I haven’t seen all of Stossel’s video, but whenever I’ve looked at these types of allegations in the past, it seems to be that those protesting that they are being suppressed are more likely to be being ignored, and primarily because they are questioning it on bases other than scientific. (IIRC in one case their work didn’t qualify on a scientific basis for admission to a conference, and they kicked up a stink. If your “science” doesn’t meet basic scientific standards, or doesn’t hold up to scrutiny and people therefore refuse to spend time taking your ideas seriously, that’s not “suppression”. That’s “science”.

    IIRC it was a BBC or mainstream UK newspaper columnist a couple of years ago who heard allegations that anti-AGW opinion was being suppressed in the scientific community. He solicited any and all information from those who felt that way, but no-one came forward with any clear evidence of suppression. I think at best there was one case looked like someone with funding made a plausible judgement call that could have gone the other way – and that’s also fairly common in science.

  147. By the way Lotharsson,

    I was thinking about your comments from yesterday on funding. Would you mind clarifying whether you think sources of funding is relevant. Or, to put it another way, where is all this funding you say is available to scientists to disprove AGW?

  148. If average global temperatures continually refuse to rise as predicted, while CO2 continues to rise, then the hypothesis is false.

    This statement as I have pointed out in the past, is a gross over-simplification that may lead to false conclusions.

    There are many systems where the overt outcomes (e.g. like “average global temperature” here) are impacted not only by known inputs (e.g. “atmospheric CO2 levels”) but also by covert processes (i.e. those whose state variables are not so easily observed – e.g. amount of greenhouse gas locked up in biomass or the sea or the like). Some systems will respond to an increase in the observed input with a temporary decrease in the measured output whilst the state of the covert processes change. When they change enough the measured output will begin to increase. These types of systems are well known in engineering control theory (and may be the result of even very simple dynamics – maybe only a couple of interacting mechanisms.) They can be difficult or impossible to predict or control unless you understand the covert dynamics.

    Climate is particularly complex. But one small example may be illustrative. IIRC (parts of) the Antarctic cooled for a while over the last few decades, even as CO2 levels increased. Does that disprove the hypothesis that rising CO2 levels leads to rising climatic temperatures? No – if you look at other processes that affect those temperatures at the same time, such as the ozone hole. The ozone hole added a certain amount of cooling to parts of Antarctica. As that hole heals, that effect will go away – and it seems very likely that those parts of Antarctica will experience warming. (IIRC it’s already being observed in parts.)

    So…you need a pretty good understanding of all the processes that affect climate in order to figure out the overall response to human activity.

    Of course Lotharsson touches on the wiggle-room that AGW proponents have worked into their theory: he calls it “noise”.

    Tony, that’s flat-out ignorant and/or disingenuous. You can do better than that.

    Noise is present in all measurements. Science training starts out very early (in some of the first science classes in early high school) learning to quantify it and take it into account – EVERY single time you do an experiment. Any time you want to disprove a theory, any theory, you have to take the measurement noise into account. Sometimes this is fundamentally harder than other times, so the debate tends to focus for quite some time on how good the measurements are and how well the noise is characterized – and so it should!

  149. Or, to put it another way, where is all this funding you say is available to scientists to disprove AGW?

    Energy companies have been particularly willing to fund research over the last few decades that might help their business avoid drastic change (although some such as BP started to invest heavily in alternative energy research recently too); so have various industry bodies and certain politically-motivated associations.

    You would also expect to get funding from the same sources as the scientists whose work tends to support AGW – those bodies don’t tend to fund work on the basis that the results would be anticipated in advance or because they fit some political or ideological or consensus filter; they usually have charters that fund work based on its value. It’s hard to argue that there would be little value in disproving AGW – especially given the anticipated costs of shifting away from a heavily-carbon-based economy.

  150. Lotharsson,

    To save us going over old ground again, and to spare everyone else the spectacle, would you at least concede this: if temperatures continually refuse to rise – through whatever cause you’d like to nominate – climate change proponents should stop using alarmist language like crisis, dangerous, and catastrophic?

  151. “It’s hard to argue that there would be little value in disproving AGW – especially given the anticipated costs of shifting away from a heavily-carbon-based economy.”

    Again, it has nothing to do with “disproving or proving” a theory once it has become “mainstream”, in fact proving or disproving becomes the after thought, as has happened…..No AGW, no funding……….simplistic yes but a well known reality in academia……

  152. Yes, Tony, with fairly strong caveats.

    If by “continually refuse to rise” you mean we don’t observe them rising for a climatically significant period of time AND we don’t observe changes in other state variables (e.g. that reflect the “covert” type of processes I referred to before) that drive processes that will significantly increase temperature – then yes. But if we see clear evidence that the underlying state is changing in ways that will drive significant temperature increase and will take a long time to control or reverse…then that may be considered a dangerous/crisis situation.

    In other words, if I’m driving a car under normal conditions and my speed is basically constant and I have enough room to stop, it’s not normally considered dangerous. If however the same is true, except that I note that the road has suddenly turned from tar to gravel and started sloping downwards to a significant degree and there’s a curve coming up…

  153. “changing in ways that will drive significant temperature increase and will take a long time to control or reverse”

    Granted we can “control” anything to begin with……

  154. Thanks Lotharsson, that’s a fair answer. The rest of it, we both know, we’re unlikely to agree on – and I, for one, am not up for another of our week-long ‘battles’ (not at the moment, anyway).

  155. Again, it has nothing to do with “disproving or proving” a theory once it has become “mainstream”, in fact proving or disproving becomes the after thought, as has happened…..No AGW, no funding……….simplistic yes but a well known reality in academia……

    You are still missing the point.

    Firstly, I agree that some work – that which looks at either the effects of GW (no “A” there), or mitigation strategies for AGW – depends on the plausibility of GW/AGW as a hypothesis for motivation. (And it doesn’t need PROOF of GW/AGW; just a reasonable likelihood.) But this work, whose funding may be in part or in whole dependent on the plausibility of GW or AGW, does not address the question of proof/disproof of AGW, and therefore its funding does not bias opinion on whether AGW is “disproved/proved” or not.

    In contrast, the work which is used to prove/disprove AGW does not – by definition – depend on whether AGW has mainstream acceptance or not. (In fact, much of the work producing evidence used to support AGW was or can be done and has value without the presence of an AGW hypothesis at all.)

    Unless you separate out those two categories of work in your thinking, you will continue to come to false conclusions.

  156. Granted we can “control” anything to begin with……

    Yes, I agree – that’s a very big “if” right now.

    I tend to think we’re at or past a tipping point already if the models are roughly right, and as a global society we haven’t shown we can exert the kinds of control we probably need to.

  157. Tony, on May 27th, 2009 at 10:43 am Said:
    “If average global temperatures continually refuse to rise as predicted, while CO2 continues to rise, then the hypothesis is false”

    Has the ice core data showing CO2 lagging temperature still valid or has it also been discredited?

  158. John,

    Has the ice core data showing CO2 lagging temperature still valid or has it also been discredited?

    I was under the impression that even AGW proponents didn’t contest that anymore, until I saw someone here the other day say it wasn’t true (although it would depend on who the commenter was, as to how seriously you take their assertions). I don’t think he said it, but you’d be better off directing that question to Lotharsson.

  159. I can’t spend much more time on this today, and I don’t know about the validity of that ice core data.

    What I do know is that arguing that BECAUSE it lagged in the past, that must prove it doesn’t drive temperature increases is not a valid argument (especially if there are certain types of feedback loops in the system, and/or more than one factor that drives temperature). [Note that I’m not arguing here that CO2 does or does not drive temperature increases – just addressing what I assume is the argument that makes reference to the ice core data.]

    Here’s a very rough analogy showing a system where the relationship isn’t that simple.

    If I drive my car and increase my fuel consumption by pressing on the accelerator, my speed goes up. If I drive it down a hill and it picks up speed, then my fuel consumption goes up (as the revs increase).

    Does fuel consumption drive speed, or does speed drive fuel consumption, or both (and do other factors also affect the relationship)?

  160. Lotharsson, on May 27th, 2009 at 12:49 pm Said:
    “If I drive my car and increase my fuel consumption by pressing on the accelerator, my speed goes up. If I drive it down a hill and it picks up speed, then my fuel consumption goes up (as the revs increase).”

    Not only rough but a really bad analogy because you always drive with your foot to the floor. The question should be if the girl sitting in your lap pushed your foot with her foot what would happen.
    The other question is what IS your relationship with the girl sitting in your lap. 😉

  161. johnd, no comment on those questions 😉

  162. Leonard Weinstein:

    In order to support a theory, specific predictions need to be made that are based on the claims of the theory, and the predictions then need to happen. While the occurrence of the predicted events is not proof positive of a theory, they increase the believability of the claims. However, if the predictions are not observed, this tends to indicate the theory is flawed or even wrong . . . .For example weather forecasting often does a good job in the very short term but over increasing time does a poor job. This is due to the complexity of the numerous nonlinear components . . . . However, for some reason, the present predictions of “Climate Change” are considered by the AGW supporters to be more reliable than even short-term weather forecasting. While some overall trends can be reasonably made based on looking at past historical trends, and some computational models can suggest some suggested trends due to specific forcing factors, nevertheless, the long term predicted result has not been shown to be valid. Like any respectable theory, specific predictions need to be made, and then shown to happen, before the AGW models can have any claim to reasonable validity.

  163. (Continued)

    Final Paragraph:

    The final question that arises is what prediction has the AGW made that has been demonstrated, and that strongly supports the theory. It appears that there is NO real supporting evidence and much disagreeing evidence for the AGW theory as proposed. That is not to say there is no effect from Human activity. Clearly human pollution (not greenhouse gases) is a problem. There is also almost surely some contribution to the present temperature from the increase in CO2 and CH4, but it seems to be small and not a driver of future climate. Any reasonable scientific analysis must conclude the basic theory wrong!!

  164. Well seeing climate dosnt affect pollution but works the otherway round id like to see our one day “clean australia day” be expanded on.

    Lets clean up our garbage first, The climate may be helped with new technology that can arise by doing things different.

    lets set our own example and not wait on what the rest of the world is doing. There is rubbish on ground and underwater, we have done so well as a nation with talk.

  165. Hi aquanut,

    I’m heading up your way tomorrow for a few days. (Don’t think I’ll have time to take that “cruise” though.)

  166. Tony
    If you like i’ll ask for Joni or Reb to pass my Email to you and catch up.

    If not a boat ride maybe its a dive you were after( i know i always come back)? 😉

  167. I know you always come back. That’s what I’m worried about. 😉

  168. Dont be worried about me.

    Tony if you happen to be on Toohey street there is a park on the corner with three houses on the opposite side, im the middle one with chooks.(would be good if you get chance)

    Im sure i have wetsuit to fit you..;)

  169. I’m not sure where Toohey St is, but I’ll be up North, on an island on the Reef. Sorry if I led you astray. (Although I would like to meet the chooks. The hen house does have an exit, doesn’t it?)

  170. false advertising…. Now i really want to take you diving.
    I have trained the chooks to attack and debone a person.

  171. Eggsactly what I was afraid of.

  172. That is not to say there is no effect from Human activity. Clearly human pollution (not greenhouse gases) is a problem.

    Curious statement. 😉

  173. Legion,

    It is ambiguous, as written.

  174. “does not address the question of proof/disproof of AGW, and therefore its funding does not bias opinion on whether AGW is “disproved/proved” or not.”

    Geez….You’re telling me that by conducting studies of any kind that look at some “cause” and/or “effect” related to AGW they are not inadvertently supporting the “theory”? Well of course they are, the very fact they are receiving funds of any sort “related” to the question is in essence supporting the theory. Funding they would not be receiving otherwise, surely you understand this……

    “In contrast, the work which is used to prove/disprove AGW does not – by definition – depend on whether AGW has mainstream acceptance or not.”

    I would agree as mentioned already but the explosion in funding in all areas related is what I am referring too, which have mentioned already. Do you think we would be interested in the nano related technological possibilities of concrete and AGW otherwise? Please…..

    “In fact, much of the work producing evidence used to support AGW was or can be done and has value without the presence of an AGW hypothesis at all.)”

    Lotharsson, I am not sure why you insist on the charades, really….There simply would not be the amount of “work” without the AGW hypothesis, as you know…….

    “Unless you separate out those two categories of work in your thinking, you will continue to come to false conclusions.”

    The only false conclusion is your belief that even without the AGW theory we would have this burgeoning industry/ grant funding etcetera that has sprung up since “Gore”, which anybody with an “unbiased” mind can clearly see for themselves………Deny all you like, but I am pretty sure “most” understand where I am coming from…….

  175. Do you think we would be interested in the nano related technological possibilities of concrete and AGW otherwise?

    Another curious question, even if it is ‘begging the question’ again and doesn’t step back from its conclusion-as-premise.

  176. One ‘we might‘, where ‘we’ is not a collection of Sparta-like I’s or even a ‘most’ of such a motley crew manning Spaceship One …

    There are many historical examples of civilizations undergoing large-scale transitions, such as the Industrial Revolution. The transition between Kardashev scale levels could potentially represent similarly dramatic periods of social upheaval, since they entail surpassing the hard limits of the resources available in a civilization’s existing territory. A common speculation[18] suggests that the transition from Type 0 to Type I might carry a strong risk of self-destruction since there would no longer be room for further expansion on the civilization’s home planet, similar to a Malthusian catastrophe. Excessive use of energy without adequate disposal of heat, for example, could make the planet of a civilization approaching Type I unsuitable to the biology of the dominant life-forms and their food sources. If Earth is an example, then sea temperatures in excess of 35 °C would jeopardize marine life and make the cooling of mammals to temperatures suitable for their metabolism difficult if not impossible. Of course, these theoretical speculations may not become problems in reality thanks to the application of future engineering and technology.

    But, as johnd so astutely points out, we’d need a controlled and bias-free experiment to test either the Spartan or the Legion hypotheses, and alas there is only one Earth.

  177. Deniers put your mouth (or typing fingers) where your heart is and argue your case here:

    A Taxonomy of Delusion

  178. Deniers??? Don’t you mean heretic.?

  179. Adrian, do you have an answer to my question of 25 May at 4.43pm?

  180. Sparta, I really don’t think you understood the problem I have with your argument.

    To summarise, your overall argument appears to have been that:
    (a) Claims are made that many scientists’ work support the position that AGW is likely to be true (and implicitly that few support the opposing position)
    (b) But many scientists whose work does not address WHETHER AGW is true, are doing work that is only relevant or of high enough priority to receive funding BECAUSE of the likelihood that AGW is true. Thus their “support” for AGW is suspect.
    (c) Therefore those scientists in (b) should not count in the numbers in (a), thus making argument (a) suspect.

    The first problem is that the scientists in (b) are generally NOT counted in (a) in the first place by those putting forward the argument. And if they are, feel free to insist on my behalf that (b) is not counted in (a). But even then, you need to assess the resulting numbers in the “correct” formulation of (a) – just asserting that it must be false because some (b)’s were counted in (a) is a fallacy. And I’m betting you’ll still have the large majority whose results supports AGW.

    Furthermore, you seem to argue that the scientists in category (b) would be out of a job if AGW were disproved, so they are motivated not to rock the boat. I’m pretty sure you haven’t worked in research, or you would know that they would simply move on to another project unrelated to AGW for which they could obtain funding. It’s not like suddenly all the scientific questions will have been answered! Without AGW, priorities for funding would shift and so would the research questions that are addressed.

  181. Oh, and Sparta, one other thing before I have to go.

    Feel free to correct me if I’m wrong, but your argument seems to rest on the idea that scientists whose current projects are of relevance because AGW is likely to be true are going to commit some form of scientific fraud or bias in order to avoid endangering their funding. In other words, if they come across something that appears inconsistent with AGW they suppress it.

    Would you elaborate on what you see the payoffs being for the different courses of action available to any scientist who finds themself in that position?

  182. Cyclones and hurricanes, and other extreme events, are supposed to increase as AGW takes hold. I can understand that as these events basically redistribute a large amount of heat in a very short space of time. I would expect that not only would they become more frequent, but more extreme, and that the “season” when they normally occur would start to stretch out, or at least shift in relation to the calender.
    To get they full picture one needs to look at all global events, but just looking a couple of indicators for our region I’m not sure what to think.
    For Australia, out of the 10 wettest cyclone, (rainfall perhaps being a reasonable indicator as to the amount of energy expended, covering that earlier period of time before more sophisticated measurements were available) “Grace” 2004 ranked 7th, “Peter” 1979 ranked 3rd, all the others being prior to that, back into the late1800’s, Mackay Cyclone 1918 ranking 1st.

    Of the most active West Pacific seasons, the top 14 years,those having 30 or more tropical storms, has 1964 with the most, (39), 1962, 66, 72, 90, 2004 each with 30 storms, all the others within the years 1962 to 1996. 13 of the worst years 1962 – 1996, only one from 1996 to 2009.

    Does anyone have links to charts or graphs that cover the global historical situation?

  183. But what about the bushfires, John, the bushfires???????

  184. Lotharsson, on May 28th, 2009 at 10:36 am Said:

    I think we should listen more to those scientists who previously supported the IPCC, but then changed their views. There surely are some valid reasons for their doing so.
    Meterologists who previously worked for BOM, but determined that the BOM modelling was flawed, and left to successfully create their own, are able to demonstrate in a very short span of time the validy of their modelling.

    More time is required for the dissenting scientists to validate theirs, perhaps in the meantime we should be paying more attention to their reasons for changing.

  185. James of North Melbourne, on May 28th, 2009 at 10:48 am Said:

    But what about the bushfires, John, the bushfires???????

    Exactly, exactly!
    Disregarding damage to infrastructure, when and where were the most destructive fires?
    Many factors should be allowed for though, access, communications, resources available, fuel reduction burns, etc, all fairly recent influencing factors.

  186. Tony, re: predictions. The climate models that climate scientists build are prediction engines, just as models are in various other disciplines. Feed them any initial state + a set of inputs and they will predict a set of outputs. (Yes, there are uncertainties involved as real world measurements are noisy and models generally fail to capture all of the interactions that affect an output.)

    Models are generally validated by testing using known historical data (Google “backtesting”). You go back in time to a starting point in the historical data set and run your model using the state at that time, plus the historically known inputs. You then compare the outputs from that time on with historically known outputs. If your model is outside the error bounds, then it has failed a prediction test. If you consistently get results consistent with history, no matter where you start the model running, then you have more confidence in it. And every subsequent real world measurement in the same series of data provides new chances to make and test your predictions.

  187. I think we should listen more to those scientists who previously supported the IPCC, but then changed their views.

    I don’t know about the (former) BOM scientists or why they left, but if their alternative models stand up to scrutiny, we should certainly pay attention to those models. And if they demonstrated reasons why the existing modeling was flawed, we should listen to them whether or not they have a better model of their own.

  188. Mobius Ecko, on May 28th, 2009 at 7:23 am

    That’s an interesting link. A link off it contains another taxonomy that I liked (and allowing for skew-ness); it’s like tick-a-box.

  189. johnd, on May 28th, 2009 at 10:57 am Said:

    I think we should listen more to those scientists who previously supported the IPCC, but then changed their views.

    But what about the energy company scientists whose whole association recently recanted against the anti-Global Warming meme and against their employers for deliberately fogging the debate?

    That has been the biggest back flip in this whole thing.

    So if you want to highlight the small handful of those who have recanted against the IPCC then how about those that have gone the other way, including the scientists who were the first to raise solar as being the driving force for global warming and then recanted.

  190. James

    Adrian, do you have an answer to my question of 25 May at 4.43pm?

    What was it? I have other stuff that takes up a large chunks of my normal life here, it’s something I’ve been called is living and working, so I can only get here in bits and squirts and don’t often go back over stuff if I’ve been away for more than a day.

  191. I wonder if “listen to them” means different things to different people? To me it means hear out their case and see if it makes sense. It doesn’t mean believe their case merely because they changed their mind.

  192. Mangled the sentence as I attempt to get out an overdue project status report and blog.

  193. Lotharsson, on May 28th, 2009 at 11:54 am

    You make it sound like a quality circle in iterative learning and engineering where the incentive is to get the model of the duplicate machine increasingly more right than wrong (by throwing as many falsifying experimental runs at it as possible) rather than a static machine.

  194. Legion, I think that taxonomy is a useful collection. PSYCH-4 seems to come into play at times…

  195. You make it sound like a quality circle in iterative learning and engineering…

    Yes, that’s what I think is going on. But don’t take my word for it – while I’ve worked on predictive modeling in other domains (albeit more biased to engineering them than researching and creating them), I’m not a climate scientist.

    But if I’m right about what’s going on, then the interesting thing, which I think is poorly understood, is that the “A” in “AGW” doesn’t so much come about from a series of experiments dedicated directly to whether or not humans have contributed to GW, as it does from inspecting what the resulting model(s) say the climate would likely be without the “A” part (i.e. by running it with inputs that have known human influences on climate subtracted).

    Or the capsule summary: the “A” part in AGW is an emergent property of our best models of how earth’s climate works.

    And yes, it’s still possible the models could be proved inadequate in non-trivial ways. And yes, there’s still significant uncertainty in the models. But when you take out the human influences the backtested predictions from the current models seem to diverge sufficiently from known history that we can be reasonably confident that humans have driven a significant contribution to GW.

    And that means anyone who thinks there’s no “A” in AGW really needs to come up with a climate model that has sufficient fidelity to known history – AND provides climate predictions for the case when the human influence is subtracted that are at least as warm as with the human influences included.

  196. Lotharsson, on May 28th, 2009 at 11:56 am Said:
    “And if they demonstrated reasons why the existing modeling was flawed, we should listen to them whether or not they have a better model of their own.”

    Lotharsson, I think we are all very aware of the El-Nino/SOI because BOM has made so much of it in recent recent, but ask yourself, how much have they mentioned the Indian Ocean in the past?
    They are starting to mention it now, gradually, but search through the newspapers and see if it was ever referred to by them until very recently. YET, one of the most basic rules for those in southern and central Australia interested in the weather was always to “Look up and look west”, our weather has ALWAYS predominantly come from the west.
    I believe they felt that once they discovered El-Nino/SOI they felt that they had all the answers and failed to consider any other possibilities. Either that, or they formed too close an incestuous relationship with their cousins on the other side of the Pacific instead of those “other” people surrounding the Indian Ocean.
    I believe also that they found validation for their models because once they felt they had a reasonable strike rate of predicting an El-Nino event, they had a very good chance of predicting a drought year because statistics showed a high % of El-Nino years were drought years. What the statistics did however show, if looked at from a different perspective, was that only a low % of drought years were El-Nino years, but you never heard them say that did you.
    It is the BOM scientists who did look at the the other perspective, and found that they could make forecasts far more accurate than BOM by including the IO and other relevant data, who left. BOM will no doubt gradually change their mindset over time as they try to incorporate the enlightenment in such a way so that nobody notices that they have gradually shifted their position.
    If such obvious factors have been left out of our everyday weather forecasts, does that not mean the same obvious factors have been left out of our climate forecasts?

  197. Lotharsson, on May 28th, 2009 at 12:10 pm Said:

    “I wonder if “listen to them” means different things to different people? To me it means hear out their case and see if it makes sense. It doesn’t mean believe their case merely because they changed their mind.”

    Totally agree.

  198. Adrian, scroll up.

  199. If such obvious factors have been left out of our everyday weather forecasts, does that not mean the same obvious factors have been left out of our climate forecasts?

    Short answer – no, because weather != climate, and weather models != climate models. (In other words, its absence from weather models it doesn’t prove its absence from climate models, but nor does it disprove it. On the other hand, I’m pretty sure you could find out about the climate models…)

  200. …anyone who thinks there’s no “A” in AGW really needs to come up with a climate model that has sufficient fidelity to known history – AND provides climate predictions for the case when the human influence is subtracted that are at least as warm as with the human influences included.

    I know, bad form commenting on yourself…but the other line of attack if you think there’s no A in AGW is to show that the current models are not sufficiently good – in other words, perhaps because some key data is skewed enough that the current testing/predictions are invalid, or the models do much more poorly backtested from a different point on the same data sets, or they leave out some human factor that – if you put it in – would change the “A” conclusion in “AGW”, or have some other flaw that means their predictions on accurate data are not sufficiently trustworthy.

    A successful attack here would not disprove “A”, but it would mean that the current confidence in “A” would be shown to be at the wrong level, and perhaps even entirely unfounded.

    Remember there are three states for any hypothesis (simplifying grossly by ignoring gradations in confidence levels) – false, indeterminate and “true” (i.e., the latter meaning currently believed true by virtue of repeated success, unless and until subsequently falsified). So this line of attack seeks to revert the “A” back to indeterminate state at the current level of knowledge, whereas providing a robust successful model where adding/subtracting the human factors doesn’t result in non-trivial extra warming would falsify the “A”.

  201. …for the purposes of the debate, can we settle on 2012 as a date where AGW is proven or otherwise. …but if temps don’t return to their pre 1998 (or 2002) northward trend, can I then confidently say it’s a crock?

    I haven’t read the argument behind that proposition, and I don’t have time right now. But as I pointed out to Tony, you can only validly do this sort of “if I don’t see this result by this date” logic if you are confident that all of your other factors (and feedback loops – especially the ones with long delays – and other more covert processes) that affect global temperatures are not changing in the meantime.

    Take a look at this comment for a particular example of this type of issue.

  202. James, to put up a crude analogy:

    A driver of a car can might confidently say as he approaches a hairpin bend that “80km/h at this spot on the road is safe” because he can normally stop before the bend from 80km/h.

    However, measuring only speed and distance is not enough. His confidence is only justified if he’s sure that his brakes are working to the required level, the road surface between that spot and the wall has adequate friction, his load is normal weight, his reactions are timely, and he doesn’t have an enormous truck pushing his rear bumper bar along. If any of those factors are impaired…

  203. Lotharsson, on May 28th, 2009 at 1:49 pm Said:
    “If any of those factors are impaired…”

    Such as the girl’s foot?

    Lotharsson, on May 28th, 2009 at 1:23 pm Said:
    “(In other words, its absence from weather models it doesn’t prove its absence from climate models”

    On the other hand, so-called “unusual” weather events are constantly being used to illustrate climate change.
    However once another previous unknown cycle, in this case the Indian Ocean Dipole, which was only identified in 1999, it’s cycles have been identified as being a dominant force driving our weather as far back as our records reach, is allowed for, the events are not so unusual, and may in fact be moderating rather than becoming more extreme.
    In an earlier post James mentioned bushfires, I’m sure by now his research has revealed that the 1800’s were far worse than any more recent fires.
    I made mention previously that in respect to droughts, again the mid 1800’s appear to have been the worst period since settlement in although official BOM records unfortunately only go back to the late 1800’s, so much of the evidence available becomes subjective.
    The point I’m trying to make, here and elsewhere is, whilst popular opinion seems to indicate our weather is becoming more extreme, all the indications that I see is that perhaps the opposite is happening.

  204. Such as the girl’s foot?

    Damn, forgot about that one 😉

    …whilst popular opinion seems to indicate our weather is becoming more extreme, all the indications that I see is that perhaps the opposite is happening.

    Fair enough – I don’t know enough to evaluate that claim.
    However, it’s dangerous to draw conclusions about global warming from that particular data point. For one thing, there are average temperatures that you can refer to in order to determine if Australia has been warming…and for another, it may be true that Australian weather is getting less extreme, whilst at the very same time global weather may become more so.

  205. Er, make that “…to draw conclusions about global warming from only that particular data point.”

  206. PILMER responds to his critics with the common sense questions they cannot seem to answer…..I think your mentioned in this Adrian, those that comment on something they haven’t even yet read?

    http://www.theaustralian.news.com.au/story/0,25197,25552775-5013479,00.html

  207. This thread is titled Hockey One, Hockey Two, maybe for good reason.
    How many Hockey Sticks would there in the past if the mathematicians, (not scientists) of the time started extrapolation the data available to them at each stage where there was a reversal in the temperature trend of the time. Nobody sees them now because of 20/20 hindsight, but they are there waiting to be drawn at every change.
    The Hockey stick theory has nothing to do with science, it is a mathematical projection. based on points plotted on an X and Y axis. Scientists, or are they still mathematicians, then try to find a theory to match the maths.

  208. What I would really like to see is the scientists produce data showing where the planets orientation presently with regards to the Milankovich Orbital Cycles (or have they been discredited also), or maybe it’s also a mathematicians job, and identify when the planet was at, or close to the same orientation, in the past.
    Without such a base I don’t think any of their projections have any credence.

  209. Common sense or red herrings?

    No one denies that the climate has changed in the past due to other reasons, it is just that those reasons seem to be missing at this point in time.

  210. Legion, Adrian, and Lotharsson,

    In their analyses and taxanomies of ‘holders of the “anti-science” position’, both John Quiggan and John Mashey fail to consider the possibility that it might be their own positions on AGW that are wrong. Others might call that important omission “anti-scientific”.

  211. Joni,

    No one denies that the climate has changed in the past due to other reasons, it is just that those reasons seem to be missing at this point in time.

    Really? Which ones would those be?

    (Hi from FNQ 8) )

  212. joni, on May 29th, 2009 at 9:28 am Said:
    “No one denies that the climate has changed in the past due to other reasons, it is just that those reasons seem to be missing at this point in time.”

    I guess we will never know if those “other reasons” were obvious at the time or only became evident in the rear view mirror.

  213. Lotharsson, May 28, 11:54am,

    Models are generally validated by testing using known historical data (Google “backtesting”). You go back in time to a starting point in the historical data set and run your model using the state at that time, plus the historically known inputs. You then compare the outputs from that time on with historically known outputs. If your model is outside the error bounds, then it has failed a prediction test.

    Plimer has a bet for you:

    Computer models using the past 150 years of measurements have been used to predict climate for the next few centuries. Why have these models not been run backwards to validate known climate changes?

    I would bet the farm that by running these models backwards, El Nino events and volcanoes such as Krakatoa (1883, 535), Rabaul (536) and Tambora (1815) could not be validated.

  214. I doubt that running them backwards would account for the climate in Australia over the 1800’s, a period characterised by the fires and drought far worse than anything seen in more recent times.

  215. Computer models using the past 150 years of measurements have been used to predict climate for the next few centuries. Why have these models not been run backwards to validate known climate changes?

    I would bet the farm that by running these models backwards, El Nino events and volcanoes such as Krakatoa (1883, 535), Rabaul (536) and Tambora (1815) could not be validated.

    Is Plimer deliberately trying to look like a buffoon? (I can’t speak to whether the models are reversible in time – some modeled processes are not – but I doubt Plimer has even considered that question.)

    When the weather forecast says rain, does he leave the umbrella at home because a volcanic eruption might drive away the rain clouds?

    I have a bet for Plimer. If he is given the task of taking the train to the airport and then flying to Los Angeles for a conference, I bet he will consult a timetable in order to plan his trip. If you buy his argument (and stretching the analogy), he will not do so – because it’s merely a prediction of what will happen, and it doesn’t account for any of the random events (crashes, breakdowns) that might occur, which in his view makes it useless.

  216. The Hockey stick theory has nothing to do with science, it is a mathematical projection. based on points plotted on an X and Y axis. Scientists, or are they still mathematicians, then try to find a theory to match the maths.

    I’m not sure that’s entirely true.

    The hockey stick is used to show laypeople who can’t understand the complexities of the science something that gives them a very rough approximation of the message coming out of the science, in a form that they can understand. Climate scientists were building climate models way more sophisticated than merely extrapolation of a single curve way before the hockey stick made a public appearance.

    You can see the problem any time the science is discussed here. Many people don’t understand how it works or what it means or how to assess it, mostly because they have a more simplistic model of the world, or like to think in black and white and are uncomfortable with probability distributions and a range of possible outcomes.

  217. In their analyses and taxanomies of ‘holders of the “anti-science” position’, both John Quiggan and John Mashey fail to consider the possibility that it might be their own positions on AGW that are wrong. Others might call that important omission “anti-scientific”.

    Others might, but they would likely be unjustified.

    You are most probably drawing an invalid conclusion from incomplete data. I would suspect that they didn’t feel the need to address the question of how they came to believe the science in an article focused on a different topic – and I would guess they addressed the science questions in many other posts. And I would further suspect behind their taxonomies they have enough evidence of behaviours that work together to avoid addressing the known science too.

    There’s a chance I’m wrong, but I would be surprised.

  218. Lotharsson,

    What about this crucial* question?

    Why have these models not been run backwards to validate known climate changes?

    *I’m sure you would agree, judging by what you say above.

  219. And Tony, I think it takes a particular kind of chutzpah (and subversion of language) to argue that agreeing with the strong scientific consensus on a scientific question is somehow “anti-scientific”. Maybe you could define what you mean by “anti-scientific”?

  220. Lotharsson,

    You seem to have missed my point. Quiggan and Mashey are calling the anti-AGW position “anti-science”, and by definition those who hold that position anti-scientific,/i>.

    Ergo, using their reasoning, the act of challenging ‘accepted’ scientific knowledge is anti-scientific.

    Quite the contrary, in my view. One would have thought that was the foundation on which science is built.

  221. Sorry people I’m really flat out at the moment and had to rush of to Sydney with an hours notice and then onto Melbourne, and only got back last night. I’m still running around and will have to scoot off again soon.

    I’ve lost the train of this thread and really don’t want to go back over the points to spend long wasted time on what appears to be basically a scientific argument being argued by all of us who are ignorant of science.

    What it boils down to is the ideological stand point and nothing will change on the most part the conservatives views on this. There are some very good papers doing the rounds as to why this is and why they are so stuck into disbelieving the science on the subject and will use non-scientific or quasi-scientific data like Plimer’s to back their ideological stance.

    Any to answer one scientific question that has been proffered a suggest those here chase up a very thorough investigation of the topic of past and present warming periods by the National Academy of Sciences of the US in 2006), which concluded;

    “The basic conclusion of Mann et al. (1998, 1999) was that the late 20th century warmth in the Northern Hemisphere was unprecedented during at least the last 1,000 years. This conclusion has subsequently been supported by an array of evidence that includes both additional large-scale surface temperature reconstructions and pronounced changes in a variety of local proxy indicators, such as melting on icecaps and the retreat of glaciers around the world, which in many cases appear to be unprecedented during at least the last 2,000 years. Not all individual proxy records indicate that the recent warmth is unprecedented, although a larger fraction of geographically diverse sites experienced exceptional warmth during the late 20th century than during any other extended period from A.D. 900 onward.”

  222. Good Adrian,

    Now where exactly do they say man is the cause?

  223. Sorry I really am rushed… that should be:

    “And to answer one scientific question that has been proffered I suggest those here…”

  224. MB – “I’ve lost the train of this thread”

    This is where it is.

    Lotharson – “NO!”

    Tony – “Black”

    Sparta – “Yes”

    Mayor – “Duh”

  225. Get off your arse find the complete paper and read it. I haven’t got time to get into a pedantic argument with you that will be never ending and will not change your already made up mind one iota, no matter what references to the science I post.

    You’re not going to believe it no matter what Tony and any of us attempting to argue the science of it here is meaningless, for like what John is doing all any of will do is cherry pick bits and pieces of the science and events that suit our stance and then lambast those from the counter argument that don’t, not matter the merit of either position.

    This is a debate of science and is being done in the background at the scientific level. The ideological debate is being waged across the public media and though it is giving an outlet for many people’s views, in the scientific realm its all meaningless.

  226. Thanks Tom, perfect summation.

    I gotta scoot again, boss is my arse for a hurry up.

  227. I’ve read the NAS panel’s findings, Adrian, and Steve McIntyre talks about it in the course of analysing the whole hockey stick fiasco at the link I provided earlier in this thread:

    The findings of these panels have been spun mercilessly. Several key findings of the NAS panel (National Research Council 2006) clearly supported us. They completely endorsed our criticism of Mann’s principal components method; they agreed that the sole use of the RE statistic was insufficient statistically and, notably, they acknowledged the MBH dependence on bristlecones and stated that strip bark trees should be avoided in temperature reconstructions.

    And the answer to my rhetorical question is: they don’t.

    It is you who is cherry-picking.

  228. Lotharsson, on May 29th, 2009 at 10:16 am Said:
    “Climate scientists were building climate models way more sophisticated than merely extrapolation of a single curve way before the hockey stick made a public appearance.”

    It always comes down to a single line.

    The past has been reconstructed as a single average line from an extremely wide range of conflicting indicators.
    http://www.windows.ucar.edu/earth/climate/images/MWP_LIA_present_NOAA.gif (the yellow is the ESTIMATED error)
    The present is being presented as a single average line from an extremely wide range of conflicting data.


    The hockey stick projection is presented as a single average line from an extremely wide range of conflicting models.

    Reality is most likely within the ranges represented. Mathematicians will always calculate the simple average, but what would the average be like if each model was weighted according to it’s reliability?

  229. Tony, on May 29th, 2009 at 9:31 am

    Yes, I adverted to ‘skew-ness’; I liked the taxonomy on the basis of its ideas as they apply to both sides of a normative fence in an established debate/dialectic. I’m also observing ‘the debate’ from one meta perspective, not participating in it; so, am not bound by ‘the science’ as object of others’ enquiries or even by the discursive use of one sense of science as applied here (more the available sciences discussed in the original Hume). As described previously, I am building an homunculus for other purposes.

  230. Yes, I adverted to ’skew-ness’

    So you did. My apologies.

  231. You seem to have missed my point. Quiggan and Mashey are calling the anti-AGW position “anti-science”, and by definition those who hold that position anti-scientific,/i>.

    Ah, I see where you are coming from – but I believe you are incorrect. They are calling certain methods of responding to science “anti-science”. They do not seem to be talking about scientists who question other scientific work on a scientific basis.

    Or as one of them put it

    The usual caveats apply, i.e., this is not about normal scientific skepticism and arguments, but where the science is denied/attacked for extra-science reasons.

    (My emphasis.)

  232. Why have these models not been run backwards to validate known climate changes?

    First answer this – what validation or falsification would running them backwards give that running them forwards does not?

    I believe Plimer is asking this question in order to disingenuously cast doubt on the models by implying that they should reverse-predict events which are clearly external to the model (see reference to Krakatoa etc).

    It’s (warning: very rough analogy) like asking why the plane timetables haven’t been validated by running them backwards starting from the end of a day when a highjacking took place, and claiming they’re no good because they didn’t reverse-predict the highjacking.

  233. The usual caveats apply, i.e., this is not about normal scientific skepticism and arguments, but where the science is denied/attacked for extra-science reasons.

    Oh, I see. My apologies to you also.

  234. Lotharsson,

    Are you saying these models can’t be run backwards? If so, isn’t that problematic for you in light of your comments on ‘validation by backtesting’?

  235. Lotharsson, surely running them backwards might demonstrate, through use of substantially known data, their accuracy better than simply running them forward using estimated data?

    When you talked yesterday about validation by backtesting, I thought that they had been validated in that manner.

  236. Some thoughts on Plimer’s article in The Australian, as linked by Sparta and Tom.

    For someone claiming in the headline (maybe he didn’t write it) that the response to his work is “vitriolic”…he’s pretty vitriolic.

    He’s adept at putting up strawmen and drawing false inferences that sound plausible:

    To demonise element number six in the periodic table is amusing.

    Who’s demonising carbon? Find me a reputable scientist who thinks carbon is not essential for our ecosystem which is filled with carbon-based lifeforms.

    So depleted is the atmosphere in CO2, that horticulturalists pump warm CO2 into glasshouses to accelerate plant growth.

    The atmosphere doesn’t have to be depleted of CO2 for additional CO2 to speed up plant growth, any more than the soil has to be depleted of nutrients before adding fertiliser will accelerate growth. By the same token the atmosphere is depleted of oxygen and our bodies of human growth hormones because athletes perform better if they’re given more of them…

    During previous times of high CO2, there were climate cycles driven by galactic forces, the sun, Earth’s orbit, tides and random events such as volcanoes. These forces still operate. Why should such forces disappear just because we humans live on Earth?

    Er, no reputable scientist is arguing these forces have disappeared. To make Plimer’s statement, you have to be ignorant of how climate modeling works.

    Why is the role of clouds ignored? Why is the main greenhouse gas (water vapour) ignored?

    IIRC:
    (a) clouds are not ignored, although last time I looked there was still pretty solid ongoing debate about modeling them.
    (b) Water vapour might have the most atmospheric volume of any relevant gas, but it is NOT the “main greenhouse gas” in terms of its greenhouse effect.

    If increased concentrations of CO2 increase temperature, why have there been coolings during the past 150 years?

    FFS, this is an ignorant rookie mistake (and I pointed out some reasons why up-thread). It shows he has no idea what he is talking about – or more likely, that he does but wants to frame the questions in a way intended to mislead most of his audience.

    In my book, I correctly predicted the response. The science would not be discussed, there would be academic nit-picking and there would be vitriolic ad hominem attacks by pompous academics out of contact with the community

    If…by “correctly predicted the response” you mean “I predicted that I would ignore all of the actual science-based criticisms and pretend that only ad hominem attacks were made”. The links Adrian(?) posted in an earlier(?) thread were pretty devastating, and IIRC were largely based on scientific objection. But he’s apparently keen on playing the “victim” card because that garners sympathy from a subset of the population, which can sometimes be turned into belief.

    …there are a large number of punters who object to being treated dismissively as stupid, who do not like being told what to think, …

    And clearly there are a number of punters who like being TOLD to think that they are too smart to be told what to think, even as they are mislead…

  237. Are you saying these models can’t be run backwards? If so, isn’t that problematic for you in light of your comments on ‘validation by backtesting’?

    I don’t think I was clear enough about backtesting.

    Let’s say we have a simple model that predicts future average peak hour traffic levels on a road. It takes as inputs the total population in a region (e.g. probably the region that feeds that road), collected every month (and might even make use of the previous month’s traffic levels as well in its calculations when predicting traffic for the current month).

    Let us imagine that we have the total population of the feeder region for every month for the last 10 years, along with the average peak hour traffic levels for each of those months.

    If we want to backtest this model, we essentially run it forwards from a known point in history and make it predict from there, and compare those predictions made using known historical state with what actually happened. For example, we may start at the very first month we have data for (i.e. 10 years ago) and use the model to predict the next 10 years, and compare the predicted 10 years of average traffic levels with the actual measured 10 years.

    And if we’re doing it right, we would also pick various other starting points, e.g.:
    – 9.5 years ago (and run 9.5 years of predictions)
    – 9 years ago (9 years of predictions)
    – 7.5 years ago (7.5 years of predictions)
    – 6 years ago (6 years of predictions)
    etc.

    If the predictions that are made using starting points in the past compare well to the corresponding actual measurements made, then we have reasonable confidence in the model (bearing in mind it doesn’t model all factors that can affect traffic levels). If they don’t, we don’t have any confidence in it.

    But we generally don’t run the model backwards (i.e. trying to predict from the current population level what last month’s traffic level would have been, and then the month before that, and so on). This doesn’t give us any additional confidence in the model over and above what we get from the backtesting procedure defined above. And depending on the form of the equations involved, it may not even be feasible as there could be multiple answers for any given month that could lead to the known state for the current month. And running backwards poses difficulties for handling external shocks to the system that are not part of the model (e.g. SARS, which may have taken a whole bunch of people off the road for a month).

  238. Tony, on May 29th, 2009 at 11:15 am Said:
    “I’ve read the NAS panel’s findings, Adrian, and Steve McIntyre talks about it in the course of analysing the whole hockey stick fiasco at the link I provided earlier in this thread:”

    It’s telling that a Statistician finds it appropriate to test what is after all simply a mathematical extrapolation.
    That is the correct application of the peer review process.
    Whilst many ordinary citizens may claim not to understand the science of climate change, most people should have a better understanding of the mathematics of it as long it is presented as such and not dressed up as something else.
    Any fool can complicate a simple matter, it takes brilliance to KISS.

  239. Lotharsson, on May 29th, 2009 at 12:59 pm.
    “like asking why the plane timetables haven’t been validated by running them backwards”

    They do. How else can they ensure feeder airlines can make connections.

    Lotharsson, on May 29th, 2009 at 1:20 pm
    “any more than the soil has to be depleted of nutrients before adding fertiliser will accelerate growth.”

    Response to fertiliser is not linear, and is optimised within a certain band.

    “To make Plimer’s statement, you have to be ignorant of how climate modeling works.”

    Who has established what our present orientation is in the orbital cycles and pinpointed previous occasions when it was the same?

    “but it is NOT the “main greenhouse gas” in terms of its greenhouse effect.”

    H2O in it’s liquid and gaseous form is 400% more effective at absorbing and dissipating heat than CO2.

  240. They do. How else can they ensure feeder airlines can make connections.

    You’ve missed the point. Do they discover reverse-predictions of highjackings, plane crashes and breakdowns when they do? Plimer was slagging off climate models for not finding the equivalent if they were run in reverse.

  241. Who has established what our present orientation is in the orbital cycles and pinpointed previous occasions when it was the same?

    I’m not sure how this question will demonstrate that Plimer knows how climate modeling works, given his other statements which reveal that he doesn’t.

  242. H2O in it’s liquid and gaseous form is 400% more effective at absorbing and dissipating heat than CO2.

    This is very roughly true, but it doesn’t make H2O more important than CO2. This is because
    (a) Most H2O dissipates from the atmosphere in a few days, whereas CO2 can persist for decades
    (b) It seems that most of the human-generated greenhouse gases are CO2, not H2O.
    (c) I haven’t seen any serious proposal for a mechanism to solve global warming by limiting the amount of H2O in the atmosphere.

  243. Response to fertiliser is not linear, and is optimised within a certain band.

    OK, bad analogy. Try the athletes instead.

  244. Quick post:

    Missing science that’s right there in front of him

    Same arguments in comments as here. See nothing changes in people’s entrenched ideological views on this, nor will change. Leave this to the scientists.

  245. Lotharsson,

    Please clarify for me whether you believe the current suite of climate models should be backtested in order to validate their predictive skill.

    (An answer couched in the same terms as the question, without reference to automobiles and autobahns, or any other analogy, would be much appreciated.) 😉

  246. Lotharsson, on May 29th, 2009 at 2:25 pm Said:

    “You’ve missed the point. Do they discover reverse-predictions of highjackings, plane crashes and breakdowns when they do? Plimer was slagging off climate models for not finding the equivalent if they were run in reverse.”

    Possiblity of increased chance of plane crashes, certainly congestion at airports.

  247. Please clarify for me whether you believe the current suite of climate models should be backtested in order to validate their predictive skill.

    Yes.

  248. Lotharsson, on May 29th, 2009 at 2:33 pm Said: ”
    (a) Most H2O dissipates from the atmosphere in a few days, whereas CO2 can persist for decades

    The quick turnaround time of H2O is it’s biggest advantage. It transports heat, disperses it, and starts the cycle all over again.
    An analogy is a fleet of small airport shuttle buses that runs on demand and a single large bus that runs only when full.

    (c) I haven’t seen any serious proposal for a mechanism to solve global warming by limiting the amount of H2O in the atmosphere.
    You miss the point, the MORE H2O cycling in the system, the more heat dispersed. It’s negative forcing at work.
    If the cooling effect gets ahead of the heating input, more of it changes state to a liquid or a solid.
    If the heating input gets ahead of the cooling effect more of it changes state to a liquid or a vapour.
    The oceans are the main buffer.

  249. I don’t think I follow your logic. If H2O (in the atmosphere) is “negative forcing at work”, then it can’t be a greenhouse gas (not in the traditional sense of the term – capturing radiant heat from the sun that would otherwise end up back in space much more quickly). And it certainly can’t be the most important greenhouse gas.

  250. Lotharsson, on May 29th, 2009 at 5:33 pm Said:

    “I don’t think I follow your logic. If H2O (in the atmosphere) is “negative forcing at work”, then it can’t be a greenhouse gas (not in the traditional sense of the term – capturing radiant heat from the sun that would otherwise end up back in space much more quickly). And it certainly can’t be the most important greenhouse gas.”

    More heat coming in results in increased evaporation which puts more water vapour into the atmosphere which then, as both vapour and clouds begins limiting the heat coming in which then reduces evaporation, which reduces vapour and clouds, which then allows more heat to come in.
    Compare an overcast night followed by an overcast day, and a clear frosty night followed by a clear day.

  251. Still don’t get it, johnd. If H2O is a greenhouse gas, then BY DEFINITION it captures heat coming in from space (unless the capturing capacity is saturated etc). So more evaporation puts more H2O in the atmosphere to capture MORE heat, not less. Unless you’re referring to the reflective (albedo) increase due to more H2O in the atmosphere which reflects some heat back into space straight away, and arguing that overwhelms the greenhouse effect?

    The night-time scenarios are not about the heat coming in, but preventing (or not) the heat that is already here (at ground level) from leaving via radiation to space. That’s very different from capturing more or less of the heat coming in from space.

  252. Lotharsson, on May 29th, 2009 at 5:59 pm

    Greenhouse gases essentially absorb heat from the surface and prevent it’s loss back out to space. H2O having absorbed heat from the surface will rise and reside in the atmosphere until conditions causes it to give up the heat, condensing and returning to the surface. Whilst residing in the atmosphere it reduces the amount of incoming radiation reaching the surface, thus reducing the evaporation and hence the amount of vapour rising leading to an increase of radiation reaching the surface. That is negative feedback.
    The night time scenario of overcast skies retains the heat during the night and blocks incoming radiation next day giving warmer than normal nights followed by cooler than normal days.
    Clear skies at night allow heat to radiate off leading to frosty nights which are always followed by warm days with little to block incoming radiation.
    Obviously it’s not quite as simple as that, there are many other factors involved making for a complex system, but weather in all its forms is all about transporting heat between points of temperature differential, and H2O apart from it’s abundance, happens to be one of the most efficient means with constant points where it changes state as it either absorbs or releases the heat. As previously mentioned each molecule is 400% more efficient than CO2 and it makes up to about 4% of all atmospheric gases whilst CO2 is only about 0.037% .
    CO2 can also exist as either a solid, liquid or gas, absorbing or giving up heat as it changes state but it requires conditions not available in the natural environment.

  253. Tony, on May 25th, 2009 at 8:08 am

    Tony, thanks for providing the link to the Steve McIntyre presentation. I finally got to read through it all.

    It has always been of concern to me that reconstructed data was built on insufficient samples from insufficient widespread sites. Add now that the procedures involved in collecting and analysing the data may also be flawed, or biased.
    After the samples are first collected and analysed, the reports then begin rising through system being incorporated into increasingly detailed reports, each become more distant from the original data, with apparently no further scrutiny, until this audit.
    By now many supposedly reputable people have locked themselves and their reputations into a position they have no choice but to defend.
    Wars have been fought, and empires lost for less, defending what initially would have been relatively minor losses of face.

  254. That is negative feedback.

    Ah, right. When you said “negative forcing” earlier you didn’t mean “negative radiative forcing”, you meant “a negative feedback loop” which works against external inputs that would otherwise drive temperature change, (IIRC) rather unlike the effect of CO2 and other greenhouse gases.

  255. It always comes down to a single line.

    You have to model climate as a set of state measurements. Which means when you make measurements for the current or the past, you have to end up with a single value for each state variable (with error bars, which you seem to be forgetting the existence of – which means that it’s not a “single line” at all).

    And your implication:

    The Hockey stick theory has nothing to do with science, it is a mathematical projection. based on points plotted on an X and Y axis. Scientists, or are they still mathematicians, then try to find a theory to match the maths.

    that the predictions are mere mathematical extrapolations of a single line are clearly incorrect.

    The predictions made by climate models are probability distributions for future state measurements. Go read (say) an IPCC Summary and note the range of estimated values for something as simple as sea level rises.

  256. The findings of these panels have been spun mercilessly. Several key findings of the NAS panel (National Research Council 2006) clearly supported us. They completely endorsed our criticism of Mann’s principal components method; they agreed that the sole use of the RE statistic was insufficient statistically and, notably, they acknowledged the MBH dependence on bristlecones and stated that strip bark trees should be avoided in temperature reconstructions.

    I’m sure I’ve said this before, but it was probably a year or two ago.

    These quotes are all true enough, IIRC. But also misleading if you don’t know the rest of the story.

    The NAS report supported the basic conclusions of Mann et al, particularly with respect to the last few hundred years, although they noted that less confidence can be held further back in the historical record (even though Mann et al clearly indicated significant uncertainties back then and noted that better data was desired, a fact lost on many critics). Furthermore, they noted that
    (a) correcting the identified issues only had a small effect.
    (b) the general conclusions of Mann et al were consistent (especially for the period of the last several hundred years) with several other methods of temperature reconstruction which (for example) did not rely on bristlecone pines.

    And IIRC Mann et al published a subsequent paper which improved their methods – and showed only small differences if you included or excluded bristlecone pines.

  257. I am still having trouble conceptualising hindcasting for climate; maybe it’s a chicken and egg thing. I can see how the machine runs forwards, with each iteration of complex variables producing a new state which leads to another new state and so on; but not how that omelette can be unscrambled in reverse. In other words, I can see how to take the inputs at a variety of points in time and run them forwards; but not how they can reliably be run backwards. Maybe it has to do with climate and the loops themselves never running backwards or as independent variables.

  258. Lotharsson, on May 30th, 2009 at 10:50 am Said:
    “(with error bars, which you seem to be forgetting the existence of – which means that it’s not a “single line” at all).”

    I haven’t forgotten them, but can you provide a link to a graph that includes them for IPCC temperature projections.
    Just as the error bars converge as past measurements progress towards the present, they must diverge for projections into the future at a similar gradient.
    The single line is nominal and for convenience only, going back or going forward. Reality is most likely contained somewhere within the range of the error bars, those ranges being as wide as what they are make any conclusions almost meaningless.
    Further, as it appears that most of the data seems to have come from a relatively small number of regions, with data from the southern hemisphere seemingly in short supply.
    Given that the southern hemisphere has a relatively large expanse of water compared to land, just the complete opposite to the northern hemisphere, further reduces the reliability of any conclusions drawn regarding the global climate past or future.

  259. Legion, on May 30th, 2009 at 11:51 am Said:
    “but not how they can reliably be run backwards.”

    I might be wrong, but perhaps running them backwards is poor terminology for climate models.
    What I think is meant is to take the model intact back to an earlier point in time and see what it throws up run forward from there.

  260. Lotharsson May 30, 11:14am,

    The following excerpt from the McIntyre /McKitrick submission to the NAS Expert Panel seems to address most of the points you raise:

    With respect to Mann et al. [1998, 1999] (MBH98-99), our most important objections [see McIntyre and McKitrick, 2003, 2005a, 2005b, 2005c, 2005d and http://www.climateaudit.org] are:

    • The study used “new” statistical methods that turned out to “mine” for hockey stick shaped series. These methods were misrepresented and/or inaccurately described in important particulars and their statistical properties were either unknown to the authors or unreported by them.

    • The reconstruction failed an important verification test said to have used in the study. This failure was not reported and the statistical skill was misrepresented both in the original article and by the IPCC.

    • Dominant weight was placed on proxies known to be inappropriate temperature proxies, along with, at best, misleading information about their impact and, at worst, actual withholding of adverse results;

    • The method of confidence interval calculation leads to unrealistically narrow confidence intervals;

    (Anyone wishing to find more detailed information on the topic of millennial multiproxy climate studies would do well to visit McIntyre’s blog, Climate Audit.)

  261. In other words, I can see how to take the inputs at a variety of points in time and run them forwards; but not how they can reliably be run backwards.

    Hindcasting or backtesting is NOT running the models backwards.

    It is running the models forwards from a point in the past.

  262. Tony,

    The study used “new” statistical methods that turned out to “mine” for hockey stick shaped series.

    I assume these methods are the ones that were studied by others in subsequently published papers. IIRC they were indeed reported to produce hockey-stick shaped series – but of very small magnitudes, nothing like the magnitude in Mann et al.

    Dominant weight was placed on proxies known to be inappropriate temperature proxies,…

    And subsequent work shows that removing those proxies (presumably the bristlecone pines) results in very little change.

    My summary is that McIntyre et al had important points to make, but their assessment of the impact of those points is heavily overstated. The key point is that if you discard the Mann et al study entirely – pretend it never happened – you get fairly similar results from a number of other methods, and you end up drawing essentially the same conclusions.

  263. Lotharsson, on May 30th, 2009 at 12:21 pm Said:

    Hindcasting is throwing your line in upstream and seeing if it drifts down and gets snagged on the same log as it did when you cast it in downstream.

  264. I haven’t forgotten them, but can you provide a link to a graph that includes them for IPCC temperature projections.

    I would expect you to be able to find this sort of thing yourself.

    Try Figure SPM.5 on p14 of this IPCC report (PDF). The lines and bars to the right of the graph show best estimates and likely ranges, respectively. The shading in the graph from the production date onwards shows +/- 1 std deviation for the averages of all studies – if you want more detail there, you probably have to dig in to the underlying studies.

    Or for a different flavour, try the AOGCM Projections of Surface Temperatures, either as Figure SPM.6 on p15 of the previous link, or at the bottom of this page. The column of graphs on the left are relative probabilities of different outcomes along the bottom (i.e. a probability distribution, which gives more information than a simple error bar). Each curve in the same graph is the probability distribution from a different forecasting method.

  265. johnd, my response to the request for IPCC temperature projection error bars appears to have got lost.

    I’m surprised you couldn’t find these yourself. Have you read any of the IPCC material that you imply is deficient?

    You could do worse than trying Section 10.5.4.6 from Chapter 10 (large PDF) of the AR4 WG1 report. It discusses the global temperature projections for 2100. There’s a reasonable amount of detail about uncertainties there and how they are derived. If you want to drill down further you’ll likely have to access the individual studies.

    Also, if take a look at the associated Technical Summary (also a large PDF). (BTW, Box TS.6 discusses Orbital Forcing and Milankovitch Cycles, which you seem to ask about from time to time.) Figure TS.28 on p72 (also found at the bottom of a much smaller webpage here) shows in the graphs in the left hand column the probability distributions for outcomes along the bottom axis (i.e. far more information than a standard error bar), with one probability distribution curve for each distinct model.

  266. Lotharsson, thanks for that, I haven’t got time to address all the issues, but the graphs on your last link http://www.global-greenhouse-warming.com/images/AR4TempSeaLevelSnowCover.jpg allow me to quickly illustrate one of the points I want to get across.

    Lets say that I accept that “the shaded areas are the uncertainty intervals estimated from a comprehensive analysis of known uncertainties (a and b) and from the time series (c).”
    Within themselves they may be acceptable, but they are different to the band of uncertainty of any forward projections.
    This is particularly relevant to any reconstructed data from thousands of years ago that are used to illustrate cyclic climate shifts.
    Say for instance we accept that today we take an temperature, say 10 degrees accurate to 0.1 degree.
    If we then refer to a temperature of 100 years ago, determined to also be 10 degrees but the degree of uncertainty then, is plus/minus 2 degrees, meaning the actual temperature then could have been 8 degrees, or it could have been 12 degrees, or somewhere in between, but nominated as being 10.
    When projecting 100 years into the future, any temperature measured then if it falls within the band 8 degrees to 12 degrees has to be considered as having zero trend.
    It is irrelevant if the temperature can still be measured within 0.1 degree, the degree of uncertainty of the PROJECTION is still plus/minus 2 degrees, that has been established at the starting point and does not alter.

  267. It is irrelevant if the temperature can still be measured within 0.1 degree, the degree of uncertainty of the PROJECTION is still plus/minus 2 degrees, that has been established at the starting point and does not alter.

    I’m afraid that if you mean what I think you mean, then I don’t agree, at least not in general. I think it’s firstly because you tend to use one term to mean another, but secondly there’s a more subtle point.

    The degree of uncertainty of the projection…is the degree of uncertainty of the projection, not of the trend. And this is not generally affected by the higher degree of uncertainty of further past measurements in the way you outlined. It is a function of the way the projection is constructed, and the inputs to the model, and their degrees of uncertainty. If those inputs are measured (say) to within 0.1 degrees, it doesn’t matter for the projection from current values if historical measurements not used in the projection can only be measured to +/- 2 degrees.

    The degree of uncertainty in the trend…well, you could argue your case for that. The problem with your argument is that you haven’t defined how you determine the trend (nor changes to it), particularly for the historic values. If the historic values have a symmetrically distributed error distribution, then (to a first order approximation in most cases) the errors don’t affect the historic trend. More specifically, and IIRC, many methods to fit curves to data sets will give you the same or substantially the same results whether you use the actual data set, or the actual data set overlaid with symmetrically distributed errors.

    If however historical values (for example) have a systemic bias…then the calculated historic trend (if you could make “error-free” measurements) would generally be different from the calculated historic trend from the values that include the errors. And that would certainly affect the difference between the trend of the past and of the projection. (And this is one reason why scientists worry about error distributions and especially about possible systemic bias.)

  268. Good. Glad that one is sorted. One was beginning to worry that press mills would never reliably be able to roll rough gauge aluminium blanks into tinfoil again.

  269. Lotharsson, on May 30th, 2009 at 1:50 pm

    Just following up previous pos.
    In Box TS.1 the teatment of confidence and likely hood, whilst they use random events such as tossing coins, rolling dice to illustrate the concept, do they allow for the fact that climate and weather is not random, but has a cause and effect. There is no reason why 1 in 100, or even 1 in 1000 floods or droughts cannot be repeated several times within a short span of time. In fact it is most likely they will be, because the climate or weather cycle provides the conditions during a particular phase of the cycle.
    When we look at cycles that we are aware of, such as the Pacific Decadel Oscillation with a 3 decade “dry” cycle and a 3 decade “wet” cycle, the theorectical likehood of an event occurring over a time line is not the same as the likelehood of an event occurring, or repeatedly occurring during a specific segment of that phase.
    Whilst we are aware of the El-Nino/ SOI and the PDO, it was only in 1999 that Indian Ocean equivalent of EL-Nino/ SOI was identified, and as yet I don’t think any IO equivalent of the PDO has been identified, but it is reasonable to expect that such a cycle will be identified given all weather events are closely related.
    This then brings me to the comments in TS.3.1.1 regarding El-Nino. They claim that the temperatures of 1998 coincided with an El-Nino, but not in 2005, these being supposedly our 2 warmest years.
    Technically, MAYBE correct, it all depends on how these events are defined. Our BOM cannot declare an El-Nino or La-Nina until the conditions have been in place for 3 months (making their forecasts next to useless).
    If we accept that 2005 was not a declared El-Nino year, where does that leave them with 2004 and 2006 which were?
    1998 may have been El-Nino year but 1997 and 1998 were La-Nina years. 1991-92, 93, 94-95 were all El-Nino years
    So what is the point or even the relevance of their assertion? This is a point they seem to have hung their hat on.
    tbc

  270. Lotharsson, on May 30th, 2009 at 10:50 pm Said:
    “And this is not generally affected by the higher degree of uncertainty of further past measurements in the way you outlined.”

    It is.
    The allowable tolerances are built in at the beginning.

    To use the sort of analogy that you like.
    If you jump in one of a 1000 “Yellow Cabs”, you know with reasonable certainty that it started it’s shift at the home of any one of a 1000 drivers. All you can assume is that it will finish it’s shift at the home of any one of those1000 drivers. As you talk to the driver you might be able to narrow it down to a suburb or even a street with an increase degree of certainty. But at this point in time, the moment the cab the cab stops, which is now, you have to allow for the fact that it could be anywhere out of 1000 homes.

  271. Legion, on May 30th, 2009 at 11:34 pm Said:

    “Good. Glad that one is sorted. One was beginning to worry that press mills would never reliably be able to roll rough gauge aluminium blanks into tinfoil again.”

    Legion, excellent example.
    The properties extended into aluminium from bauxite give us the aluminium ingots today, and whilst rolling it into the thin foil of today might be the trend, those properties inherent in the aluminium allows for a much wider range of possible uses in the future, some perhaps yet unknown, but all directly attributable and traceable back to the bauxite when eventually revealed.
    The possible uses were determined when the bauxite was formed and not confined by the customer demand of today.

  272. sorry typo.
    “1998 may have been El-Nino year but 1997 and 1998 were La-Nina years. 1991-92, 93, 94-95 were all El-Nino years”

    should read 1997 and 1999 were La-Nina years.

    A little bit confusing as I believe the IPCC are using calender years whereas data I refer to covers the period April to March which is more appropriate for the rainfall and growing seasons, at least in Australia.
    So to be more specific, it was 1996-97 and 1998-99 that were La-Nina years, (April-March)

  273. Lotharsson, TS page 37. Global map showing surface temperature trend. Strange that the polar regions are “uncertain” even more strange that the southern “uncertain” area is bounded by areas of zero or decreasing temperature trend.

    Other problem I see is that the assumptions are generally linear. Particularly whilst they produce graphs over short time frames, many seem to start about 1960, many even less, 1988 for some. When these refer to rainfall or temperature, these starting points should not be used because they fail, (or perhaps not) to take into account the point at which some longer term weather cycles are at.
    They seem to jump all over the place with “trends” established over very different time frames.

    Then there is the matter of Greenland.
    Much about what has happened in the last couple of decades, but why don’t they refer to the conditions there in the first half of the 1900’s? Is this omission deliberate, because including it would negate many of their conclusions based on the last couple of decades.

    Granted this is a big subject, difficult for any person to get their minds around. Perhaps even for the experts with a deep knowledge of a very narrow field.
    At the end of the day it may be the ordinary person who is standing several steps back, taking in the broader view, evaluating what is being said against what is being observed, that as a group may have a better idea than the experts. That also is part of the peer review process.

    It certainly works at football matches, and it works on the stockmarket, and it works in politics.

  274. There is no reason why 1 in 100, or even 1 in 1000 floods or droughts cannot be repeated several times within a short span of time.

    This is also a property of purely random processes such as coin tossing, but I take your point. (If you toss a coin 1,000,000 times and DON’T see a run somewhere of about 10 heads in a row somewhere in that 1,000,000 events, you’d start to get concerned about the randomness of your coin toss.)

    If it’s a climate effect (longer term trend), rather than a weather effect (short term occurrence), then I would be surprised if it were not evident in the models. I imagine that in most models it emerges as a property of the model of underlying conditions (roughly speaking “cause and effect”) in most cases, than a specific model of the types of oscillations you are referring to.

  275. It is.
    The allowable tolerances are built in at the beginning.

    I don’t know what you mean by “allowable tolerances”, so I am unable to understand your argument here.

    But try this example.

    Ever done high school physics? One of the first experiments you might do is to measure the speed of a little cart running down a slope as it accelerates under the influence of gravity.

    When I did it, we used a “ticker tape”. The cart dragged a piece of paper behind it, which ran through a little gate where a metal pointer banged up and down at a fixed frequency. Whenever it banged on the paper it left a mark. That way you could figure out

    average speed = distance / time

    during every banger interval.

    Suppose you built a little model that predicts the future speed of the cart. If you know the current speed (and the constant acceleration), you can predict

    future speed = current speed + acceleration * time interval

    Note that the only model input is the current speed. Acceleration and time interval are parameters.

    Now we haven’t dealt with errors yet. Imagine that the speed measurement via ticker tape has an error bar of +/- 10%. Imagine you also had an optical distance detection system hooked up to a timer which let you calculate speed with an error bar of +/- 0.1%.

    Now, built yourself a nice long slope. Start the cart at the top, and measure its speed using the ticker tape. Part way down, switch to measuring speed via the optical system. Now use your model to predict the future speed after a suitable additional time interval.

    Leaving out the model’s own error bars (which should be the same either way), which input value (i.e. speed measurement) error bars influence your model’s predictions? The +/- 10% or the +/- 0.1% or both? Why or why not?

  276. It certainly works at football matches, and it works on the stockmarket, and it works in politics.

    I’m not sure what you’re getting at there, but even if “it works” for those things, the analogy is deeply flawed. You need to be very careful about arguing from analogy, rather than using them to illustrate.

    Scientific truth stands whether the crowd believes it or not; furthermore the truth of any straightforward hypothesis is not directly and immediately influenced by the actions of the crowds. This is very unlike politics, which (in elections and opinion polls) is ALL about what the crowd thinks, or the stockmarket which is all about what the market players do.

  277. The properties extended into aluminium from bauxite…
    The possible uses were determined when the bauxite was formed and not confined by the customer demand of today.

    That is a very strange way to think of it.

    The properties of aluminium ARE the properties of aluminium. Those properties are not derived from those of bauxite, even though aluminium itself is.

    The possible uses of aluminium are defined by the properties of aluminium. They were not in any way “determined when bauxite was formed”.

    If your argument held true, you could say that the properties extended into diamonds and graphite from CO2 gives us the excellent cutting applications of diamonds and the writing & lubrication applications of graphite and today, and those applications were determined when the CO2 was formed…which doesn’t make any sense. The properties of CO2 do not give any direct insight into the properties of different forms of C.

  278. Other problem I see is that the assumptions are generally linear

    Do you see declarations anywhere that the “assumptions are linear“, or do you see graphs that appear roughly linear over some portion and infer that means that the assumptions are linear? Such an inference is likely unwarranted. Many non-linear physical systems exhibit near-linear behaviour within certain regions, or over relatively small areas. But that doesn’t make them linear systems.

  279. …do they allow for the fact that climate and weather is not random, but has a cause and effect.

    I think this is worth coming back to.

    You follow on from this comment to discuss various weather patterns and oscillations, and I get the feeling that you think these oscillations are the cause and effect that you refer to, and that without modeling these oscillations themselves the results may be invalid.

    It’s also theoretically possible to model the underlying physics and chemistry that causes the oscillations you talk about. In other words, the oscillations are an effect of something else, and are produced by the model which also then produces the associated climate effects.

    You’d probably have to dig into the models themselves to find out how they’re built.

  280. Lotharsson, on May 31st, 2009 at 12:02 pm

    Both, at least. Perhaps compounded.
    The “current speed” at which the more accurate timer took over is still only known to within plus/minus 10%.

  281. Lotharsson, on May 31st, 2009 at 12:21 pm Said:

    “It’s also theoretically possible to model the underlying physics and chemistry that causes the oscillations you talk about. In other words, the oscillations are an effect of something else, and are produced by the model which also then produces the associated climate effects. ”

    Yes. Whilst we are gradually getting an understanding of the cycles of El-Nino/SOI, and the ongoing effects of such an event happening, we still don’t know what triggers them. The PDO seems to be one longer term cycle that increases/decreases the likelihood, but is the PDO only part of an even longer cycle?
    Our knowledge is still extremely limited in many ways. As mentioned earlier the criteria which allows BOM to declare an El-Nino requires the conditions to have been in place for 3 months before they can do so. Then if it is correct they claim a successful forecast. If a drought results then they claim another high strike rate of El-Nino being the cause of droughts.
    However if you work it backwards starting with the drought years, only a low % of drought years are El-Nino years. I think BOM have woken up now taking notice of the research of others, but they still haven’t refined their models to match those of private forecasters.
    Last year, one Monday, BOM and another forecaster issued an outlook for the next month that were the exact opposite of each other. By the Friday BOM had adjusted their forecast to match the other one.

  282. Both, at least. Perhaps compounded.
    The “current speed” at which the more accurate timer took over is still only known to within plus/minus 10%.

    That is flat out wrong. Until you understand this point you will continue to draw false conclusions.

    Take a look at the equations again.

    The ONLY input that affects the predictions is the current speed. If, at the point in time that I make the future speed prediction, I measure the current speed to +/- 0.1%, it makes no difference to the prediction that I may (or may not) have measured speed at an earlier point in time with a different accuracy.

  283. However if you work it backwards starting with the drought years, only a low % of drought years are El-Nino years. I think BOM have woken up now taking notice of the research of others, but they still haven’t refined their models to match those of private forecasters.

    Fair enough, but that is still a discussion about weather forecasts, not climate modeling.

  284. johnd, let me riff off your (flawed, but perhaps still useful) cab analogy and see if the lightbulb goes on.

    Some years ago there were 10,000 cabs in the city you mentioned. Now, due to hard economic times, there are only 1000. When you get in the cab and ask yourself how many possible homes there are for that cab to end up in at night, do you think 1000? Or 10,000? Or 1,000 somehow “compounded” with 10,000?

  285. The “current speed” at which the more accurate timer took over is still only known to within plus/minus 10%.

    And just to make sure we understand each other here…the “current speed” at the instant the more accurate timer made its first measurement is known to +/- 0.1%. The more accurate timer “knows” nothing of the less accurate timer, and couldn’t care less if another measurement was made before, at the same time, or after its own measurement(s) were made.

  286. Lotharsson, on May 31st, 2009 at 2:28 pm Said:

    “The “current speed” at which the more accurate timer took over is still only known to within plus/minus 10%.
    And just to make sure we understand each other here…the “current speed” at the instant the more accurate timer made its first measurement is known to +/- 0.1%. The more accurate timer “knows” nothing of the less accurate timer, and couldn’t care less if another measurement was made before, at the same time, or after its own measurement(s) were made.”

    Lotharsson,
    I’m glad you added this post, because I was going to reply to your earlier post, but this makes it easier. I think you might have mentioned something about an electrical background so you should see the relevance of this point.

    The accuracy of the timer may be 0.1%, BUT where have you calibrated it, and then zeroed it?

    You cannot say that you calibrated it elsewhere because in the context that this analogy, all you have as reference points is the data collected earlier with the less accurate timer.
    You have to tell the new measuring device that if it measures X distance in Y time then that equals S speed, it cannot work that out without input from you, and as your analogy indicates, your input data is only accurate within 10%. So in reality what the machine reads as say 100, could be anything from 90 to 110 because that is all you know for certain, and that error is then inbuilt. The more accurate timer HAS to have “knowledge” of the less accurate timer because that is where the data it is programmed with originated from.
    I know what you are going to say is, if that was the case accuracy would never improve. Accuracy does improve all the time, but it cannot be proved in advance. Only after data has been collected and then compared to known standards can we determine the accuracy of the new method, but that is after the fact, and in real terms it always comes back to what are our original benchmarks.
    That is more or less the point I was trying to make with regards to climate projections

    Take weigh bridges as an example. Whilst they may accept weights of hundreds of tonnes, they are calibrated using much smaller weights, perhaps one tonne. The test weight is of a known accuracy. You start off telling the weighbridge that its weight is zero, even though it actually weighs 100 tonnes. Then you put a load on say about 20 tonne. The bridge then tells you it thinks it is 22 tonne. You accept that and put the one tonne test weight on and tell the bridge that it now has 23 tonne on. The bridge then works out that the only thing it knows for sure is the difference between 22 tonne and 23 tonne. This goes on using progressively bigger loads, 50 tonne, 100 tonne, 200 tonne, and so on, all the time adding a roughly known load and then the accurate one tonne weight. At the end the bridge has accurate data for the ranges 22 to 23 tonnes, 50 to 51 tonnes, 100 to 101 tonnes, 200 to 201 tonnes and so on. Using that data it spans it across the full range, using what you told it was zero as the starting point. However if your test weight had a 10% error, then despite the bridge being capable of greater accuracy, it simply won’t be.
    A bit long but I think you will see what I’m getting at.

  287. Lotharsson, on May 31st, 2009 at 2:24 pm Said:

    “Fair enough, but that is still a discussion about weather forecasts, not climate modeling.”

    Is it possible to define climate without reference to anything weather related?

  288. You cannot say that you calibrated it elsewhere…

    You most certainly CAN say you calibrated it elsewhere – even if you personally can’t imagine how this would be achieved. One way is to use a known standard distance and known time interval – analogous to your one tonne known “test weight”.

    At the end the bridge has accurate data for the ranges 22 to 23 tonnes, 50 to 51 tonnes, 100 to 101 tonnes, 200 to 201 tonnes and so on.

    No, it does not.

    You could argue it could provide some accurate readings – but only for the difference between two items that read 22 and 23 tonnes, and 50 and 51 tonnes, and so on.

    But a weighbridge that reads “22 tonnes” when given a 20 tonne load is useless – it’s supposed to tell you what the load itself weighs, not the difference between two different loads.

    How do you actually calibrate it? By using a whole series of accurate test weights.

    Back to climate models. The more recent temperature measurement methods are clearly of higher accuracy than the historical proxy measures, if only because we can actually measure today’s temperature with very high precision, and have been doing so for some time at various weather stations. That means the climate models use high precision current temperature inputs, and their predictions do not rely on less precise historical measures.

    Yes, you have to be very careful to figure out how to calibrate the historical proxies – that’s often a fertile area of scientific debate. And inaccuracies there (especially if they lead to a bias rather than a symmetrically distributed error) can distort comparisons between different time periods, or between the past and the predictions of a model. But none of these things affects the error bars for the predictions made from models because they generally don’t use historical data as inputs.

  289. Is it possible to define climate without reference to anything weather related?

    No. But weather models clap out over much shorter periods than climate models, so drawing conclusions about how climate models work from how weather models work is fraught with danger.

  290. Lotharsson, on May 31st, 2009 at 11:37 pm Said
    “But a weighbridge that reads “22 tonnes” when given a 20 tonne load is useless – it’s supposed to tell you what the load itself weighs, not the difference between two different loads.
    How do you actually calibrate it? By using a whole series of accurate test weights.”

    With all our analogies I’m starting to forget what the original subject was, but never mind.
    How do you actually calibrate a weighbridge? (I shouldn’t have to mention it, but this obviously refers to electronic scales, not mechanical ones)
    You do so by using a relatively small test weight of known accuracy, such as the 1 tonne weight mentioned, and a number of loads of approximate weights to provide a series of calibration points across the full range of the bridge, say 25%,50%,75%,100%. These weights don’t have to be known at all, a guess is good enough. The calibration is done by programming the bridge first to recognise zero, and then to recognise the DIFFERENCE between each load and each load with the test weight added. The actual weight of the load is largely irrelevant except to provide reasonably regular check points across the full weight range. Once that data measuring the DIFFERENCE has been collected by the bridge, the bridge then extrapolates the data that accurately represents 1 tonne at various intervals, to span the full measuring range, in other words, it fills in the blanks as you do when creating a graph, because that is what it is doing, creating an electronic graph. It creates a best fit gradient. You have already told it what zero looks like and it takes it all from there.

    Just as with your electronic timer. If the very first measurement indicates 100, the only reason it tells you that is because you had previously told the unit, that is what 100 looks like.
    But how did you know what 100 looked like?
    Because you ticker tape timer told you so, but as you know that could be 90, or it could 110, so even though the new timer is accurately measuring what it sees, when it tells you it is 100, in reality it could be 90 or it could be 110.

    So put put this into a climate context, when our accurate measurements tell us that a new record has been set, how do we know that?
    Because our old records told us so.
    BUT how accurate were the old records?
    Was what we set as the “zero” really zero?.

    As the McIntyre audit showed, many of the benchmarks may not be as accurate as first claimed, and given the limited number of samples from a limited number of locations, not all that representative.

    You can see the subtle changes continually happening, AGW becomes Climate Change. Global Climate Change now starts to be more specific in that actually they are now referring to regions.

  291. Lotharsson, on May 31st, 2009 at 11:41 pm Said:
    “No. But weather models clap out over much shorter periods than climate models, so drawing conclusions about how climate models work from how weather models work is fraught with danger.”

    That supports a point I have been making ever since climate change became an issue.
    If we recognise decadel cycles that determine our weather, how can climate models use data from a shorter time frame, especially more recent data when the current cycle is still in progress.
    Looking for trends with starting points 50 years ago or less is meaningless, unless they completely understand the workings of the current weather cycle.
    But they don’t. far from it.
    I’ve seen projections made about Australian rainfall with the 70’s as the starting point. Naturally a couple of years ago it showed a downward trend because overall the 70’s provided one of the wettest periods on record. Make the starting point the 60’s, an exceptionally dry period, a different trend emerges. Do the same projections at the end of this year and there will be 2 different results again.
    Starting any rainfall related study at any time post WW2 is almost fraudulent. The first half of the 1900’s were drier, and the 1800’s perhaps drier still.
    So where should climate modelling, especially for Australia start from? Certainly not from the point when supposedly more accurate measurements became available.
    As a matter of interest, there is always differing opinion as to when which was Australia’s worst drought. Experts seem to want us to believe the most recent one. Personally I believe the mid 1800’s, but drought is hard to quantify.
    But here is an easy one. When did Victoria experience our worst bushfires? That should be easy, all the experts have deemed the latest one.
    Would you believe February 1851? By far the worst ever experienced. It’s all documented in the newspapers archives of the time.Weather conditions were atrocious.

    So are the most recent claims ignorance or fraud? Either way we shouldn’t let those people anywhere near any modelling concerning the future.

  292. “So put put this into a climate context, when our accurate measurements tell us that a new record has been set, how do we know that?”

    Lotharsson,
    Just to give an example that can be related to the model timer.

    If an old temperature record high was set, say 50 degrees, but the accuracy at the time was plus minus 10%, for a new record to be set, if the accuracy of measurements are now 0.1%, only when the temperature reaches 55.1 degrees can it be legitimately claimed as a new record, (actually 55.055 degrees).
    Do you agree?

  293. You do so by using a relatively small test weight of known accuracy, such as the 1 tonne weight mentioned, and a number of loads of approximate weights to provide a series of calibration points across the full range of the bridge, say 25%,50%,75%,100%

    Saying it does not make it so. It is still flat out wrong, especially:

    The actual weight of the load is largely irrelevant…

    The problem with your method is that the extrapolation of which you speak cannot be performed, because you don’t know WHERE your sampled points are. Yes, you know where your readings of “22” and “23” are relative to one another – but you don’t know how far “22” is from “0”, or from “50”…

    But how did you know what 100 looked like?
    Because you ticker tape timer told you so, but as you know that could be 90, or it could 110, so even though the new timer is accurately measuring what it sees, when it tells you it is 100, in reality it could be 90 or it could be 110.

    No, no, no.

    You know what 100 is like because you calibrated it from some other standard. Instruments with a given level of precision are not calibrated from instruments with a lesser level of precision. It goes the other way. Your earlier rambling paragraph trying to claim that higher precision can be derived from lesser precision failed to explain how that might be so.

    In particular, how does the manufacturer know the measurement precision of an instrument? They measure its performance against a more precise standard.

  294. So put put this into a climate context, when our accurate measurements tell us that a new record has been set, how do we know that?
    Because our old records told us so.
    BUT how accurate were the old records?
    Was what we set as the “zero” really zero?.

    That’s where your argument holds water, not because “what we set as zero may not have been zero”, but because of the higher error bars for more historic measurements.

    But you must understand my key point – this STILL does not affect the error bars on a model projection using today’s far more accurate measurements. Your argument has persistently confused the two.

  295. As the McIntyre audit showed, many of the benchmarks may not be as accurate as first claimed, and given the limited number of samples from a limited number of locations, not all that representative.

    I would be careful not to believe McIntyre and his erstwhile partner McKitrick only on their say-so. You’d want to see what others have to say in response first. I’m sure you can find more interesting feedback if you look.

  296. You can see the subtle changes continually happening, AGW becomes Climate Change. .

    Rubbish.

    Climate change refers to the OBSERVATION that our climate has been changing. Hardly anyone argues with a straight face that this is not occurring, and climate scientists have been looking at this for quite some time.

    AGW refers to the hypothesis that human activities have contributed to climate change. The term “AGW” has not mutated into “Climate Change”; if anything the observation of “Climate Change” preceded the widespread acceptance that humans had something to do with it.

    Global Climate Change now starts to be more specific in that actually they are now referring to regions

    They – as in scientists – have ALWAYS been referring to regions in their work. They have never maintained that every region will experience exactly the same type and amount of climate change at the same time.

  297. So where should climate modelling, especially for Australia start from?

    And you’re also concerned heavily with records and whether we can be sure about having exceeded the old ones due to less accurate historical measurements.

    This is consistent with my observation that you seem to be under the impression that climate modeling is merely an extension of weather modeling, which leads you to all sorts of conclusions that may not be accurate. Go talk to some scientists about how they model climate.

    Climate is not about individual events; nor about individual years. It is primarily about long term averages. This is why larger historical error bars may not matter as much as you think (provided that the measurement method has a symmetrical error distribution, and that careful calibration for temperature reconstructions is performed) – because long term trend calculations by their very nature tend to cancel out most of the errors. And this is why whether this year was a record or not for (say) peak temperature doesn’t matter that much to scientists either. It’s more of interest to non-scientists for whom the science is usually greatly simplified, and those arguing that the simplified version is wrong – on complex grounds.

    Speaking of individual events which should not on their own be held out as proof of climate at the time:

    When did Victoria experience our worst bushfires? That should be easy, all the experts have deemed the latest one.
    Would you believe February 1851? By far the worst ever experienced. It’s all documented in the newspapers archives of the time.Weather conditions were atrocious.

  298. So where should climate modelling, especially for Australia start from?

    Go and talk to the climate modelers, or read about their models.

    I think you’ll find they tend not to do models like “here’s a 45 year Australian rainfall cycle and we’re in year 17” and “here’s a 180 year Australian inland temperature cycle and we’re in year 146”. That doesn’t let them predict or detect any change; merely re-iterate what happened in the past. It would be a lot like your idea that the hockey stick curve was merely mathematically extrapolated and the extrapolation was called a “climate projection”.

    I expect you’ll find they mostly model the underlying physics (and perhaps some chemistry) and they initialise their models with the current measured conditions, not with historical data. That distinction may be the source of a lot of your confusion.

  299. Lotharsson, on June 1st, 2009 at 10:17 am Said:
    “Saying it does not make it so. It is still flat out wrong, especially:”

    All I can suggest is you ask someone who is familar with calibrating weighbridges, (other than me that is)

    “The problem with your method is that the extrapolation of which you speak cannot be performed, because you don’t know WHERE your sampled points are”

    What the bridge does know, because you told it, is that the measured difference in output from the loadcell, say volts, represents 1 tonne. Say the first reading was 6 volts, after 1 tonne was added, it becomes 7 volts. Because you told it 1 tonne =1 volt, the bridge therefore assumes that 6 volts must equal 6 tonnes at that point. Adding increasingly bigger loads establishes whether the 1 tonne = 1 volt relationship holds true or changes.
    The bridge also knows how many volts = zero because you also told it.

    “No, no, no. ”

    You last post answers this.
    The analogies always had be seen in the context of weather or climate matters.

    My post about the old temperature record reflects the point I was trying to make all along.
    ( johnd, on June 1st, 2009 at 6:57 am Said:)
    I assume you agree with it given you last post.

    Are there any other loose ends that need tidying up? 🙂

  300. Lotharsson, on June 1st, 2009 at 10:42 am

    Basically what I am saying, is that, particularly in relation to Australia, are the experts or otherwise who see or use extreme weather events, and including them in their calculations, graphs and presentations, (and I include BOM in there), sure that these are unusual or extreme events, and not the normal. Just because they starting record keeping 100 years ago that then should be the benchmark.
    Until we are really sure what is normal and what is not, we cannot determine if anything is changing, or if it is, in what direction or at what rate.
    I don’t think we can really tell the difference between a hot day in May or a cold day in September.

  301. Lotharsson,
    I see that BOM and the CSIRO are intending to replace their existing system of providing forecasts based on statistics with a new “dynamic” forecasting model.
    They are waiting on new super computers, but they think it may still be several years before they “could produce reliable usable predictions”

    I wonder how come those private forecasters with limited resources can do so well, I doubt they have super computers.
    I suspect that BOM and CSIRO will take the opportunity to include much more relevant data into their models, such as IO data, which alone will improve their strike rate on it’s own, and then use the improved results to validate their new approach and justify the expense.

  302. Because you told it 1 tonne =1 volt, the bridge therefore assumes that 6 volts must equal 6 tonnes at that point.

    Right, and unless you have some other means to believe that 6 volts equals 6 tonnes, you are drawing a false inference. That means might be a characterisation of the degree of linearity of the device by the manufacturer – but even THAT claim has to be substantiated, and should be checked from time to time.

  303. “No, no, no. ”

    You last post answers this.
    The analogies always had be seen in the context of weather or climate matters.

    I fail to see how my last post (before that comment) answers that. You have a fundamental misconception, and I see no evidence you have understood that yet. It is exemplified by this:

    Until we are really sure what is normal and what is not, we cannot determine if anything is changing, or if it is, in what direction or at what rate.

    We certainly can. We don’t do climate predictions based on the most extreme events observed in an interval. Climate is about averages. Understanding that fact is crucial.

    And as I have pointed out, even with larger error bars in the past, you CAN determine whether there is a trend in the long term averages provided you understand the error distribution and do a good job of calibrating historical measurements.

  304. When did Victoria experience our worst bushfires? That should be easy, all the experts have deemed the latest one.
    Would you believe February 1851? By far the worst ever experienced. It’s all documented in the newspapers archives of the time.Weather conditions were atrocious.

    ALL the experts, and just which experts in what field?

    John, haven’t you stated that past weather measurements were inaccurate meaning they were unreliable for modelling, so just what was the accuracy of the measurements in 1851? Also what was the extent in both frequency, range and spatially of the measurements taken back then, or was it as I suspect more anecdotal from witness or second hand accounts. The recent fires had satellite and aerial observation (using the Firescan system I once worked with) along with the latest instruments measuring it, whereas the 1851 fires only had rudimentary instrumentation and observation methods, so how can it be said 1851 was worse than the recent fires or vice versa?

  305. I wonder how come those private forecasters with limited resources can do so well, I doubt they have super computers.

    Again, you’re talking about weather predictions (i.e. for a given year), not climate (trends in averages over a few decades or longer).

    And do you actually know the private forecasters are doing so well? Have their predictions been adequately backtested, or is there a chance they have done some curve & cycle fitting that happens to have worked well for a little while but may not work so well in the future? (The latter type of “model” is a professional hazard in the finance industry. Most punters, and even a bunch of industry professionals, don’t have the skill or the data to figure out whether this is the case, and tend to fall prey to the (very carefully worded in fine print) marketing claims.)

    If they do predict well, it would not surprise me if the BOM picked up some of their (or similar) techniques.

  306. Are there any other loose ends that need tidying up? 🙂

    I can’t see us getting much further. You clearly have an active and interested mind and are certainly thinking about the topic. However, from my perspective, you need to gain some crucial insights and some new data, shed some cherished false assumptions, and avoid leaping to unsupported inferences. I’ve done about as much as I can to point out where and why.

  307. Lotharsson, on June 1st, 2009 at 12:09 pm Said:
    “you are drawing a false inference. That means might be a characterisation of the degree of linearity of the device by the manufacturer – but even THAT claim has to be substantiated, and should be checked from time to time.”

    Correct inference if you please.
    The progressively larger loads during calibration are the substantiation and determines the actual characteristics of the specific unit in the specific application. Periodic calibrations correct any drift.

  308. Lotharsson, on June 1st, 2009 at 12:13 pm Said:
    “We certainly can. We don’t do climate predictions based on the most extreme events observed in an interval. Climate is about averages. Understanding that fact is crucial.”

    Yes agreed. New extreme highs and new extreme lows don’t neccessarily change the average.

  309. The progressively larger loads during calibration are the substantiation and determines the actual characteristics of the specific unit in the specific application. Periodic calibrations correct any drift.

    If your accuracy relies on linearity across the full scale, then SOMEONE must verify it, because piecewise linearity does not guarantee full-scale linearity. (Even well-known curves can be closely approximated by straight lines over small ranges.)

    So you either ASSUME sufficient linearity – in which case you are relying on both the manufacturer’s assertion, AND that the assertion will hold over time – or you must VERIFY it with known loads.

    If you rely on the manufacturer’s assertions, then you’re merely passing the buck. The manufacturer still has to do a whole bunch of work with a whole range of accurately known loads in order to prove sufficient linearity – and even more work to prove that the stated degree of linearity holds both over time and the full range of operating conditions.

  310. johnd, on June 1st, 2009 at 12:32 pm Said:

    Yes agreed. New extreme highs and new extreme lows don’t neccessarily change the average.

    Yet isn’t the climate average trending upwards as are global weather anomalies?

  311. Mobius Ecko, on June 1st, 2009 at 12:17 pm Said:
    “so just what was the accuracy of the measurements in 1851? Also what was the extent in both frequency, range and spatially of the measurements taken back then, or was it as I suspect more anecdotal from witness or second hand accounts. The recent fires had satellite and aerial observation”

    Granted, but it cuts both ways. If earlier data is not accepted to support one assertion, it can’t be used to counter another.
    FYI, the fires of 1851 burnt about 5 million hectares. Data collected at the time allowed this to be broken down into the areas of various types of terrain burnt, bush, grasslands, farmlands etc. documented in official journals, recording the number of lives lost, as well as settlements, individual buildings, sheep, cattle, etc. All in all, probably fairly accurately accounted for, but it was fairly quickly forgotten about probably due to the gold rush which took everybody’s interest shortly after the fires.

    There is an unknown amount of recorded information, station journals etc., preceding official records, residing at old established settlements and farms, or in family archives, and increasingly in local museums. Periodically something new comes to light and provides an detailed insight into events and conditions of a specific location, often confirming other similar revelations, but generally it remains of local interest. What was collected by the authorities at the time is often buried deep in official archives and paid scant attention to today.

  312. Mobius Ecko, on June 1st, 2009 at 12:36 pm Said:
    “Yet isn’t the climate average trending upwards as are global weather anomalies?”

    Yes, but have they moved outside the range of uncertainity established by the historical data???

  313. Lotharsson, on June 1st, 2009 at 12:36 pm Said:
    “If your accuracy relies on linearity across the full scale, then SOMEONE must verify it, because piecewise linearity does not guarantee full-scale linearity. (Even well-known curves can be closely approximated by straight lines over small ranges.)”

    Exactly, a curved line is a series of straight lines, the number and length of each depends on your required accuracy.
    I think you might be about ready to do your first bridge calibration 😉
    But don’t forget.
    The known load is the test weight,
    and don’t forget to zero it. 🙂

  314. Granted, but it cuts both ways. If earlier data is not accepted to support one assertion, it can’t be used to counter another.

    No, that does not hold. There is a crucial difference that has been explained before. Uncertainty about measurements of a single extreme event is markedly different from uncertainty about averages derived from a series of such measurements.

    Furthermore, you can’t point to how many hectares it burnt as a valid comparison unless you’re prepared to argue that firefighting methods were as effective then as now, that combustion loads were at similar levels per unit area, bush composition is similar to today, and perhaps even that the area of bush available to burn was similar.

  315. Yes, but have they moved outside the range of uncertainity established by the historical data???

    Well, that’s the fairly clear conclusion (depending on exactly what you mean by “historical data”) of the IPCC.

  316. Exactly, a curved line is a series of straight lines, the number and length of each depends on your required accuracy.

    That may be true, but your procedure does not calibrate such a curve.

    If your transducer gives you a curved line, then by definition it is not full-scale linear. If that’s the case, your calibration process is fatally flawed. You don’t know WHERE on the curve your pairs of points separated by a known load lie, therefore you can’t interpolate. It’s not that hard to understand.

  317. There is one other factor that makes measuring past events today more accurate in that with the forensic technologies, resources and new knowledge available nowadays scientists can more accurately measure a past catastrophic event, even ones that occurred centuries ago, than could be measured when the events actually occurred, that is if they were measured at all at the time.

    Forensics of the 1851 bush fires done today would reveal more about their ferocity and extent than witness accounts of the time, and they could be directly compared to bush fires since then up until now. This is quite legitimate and goes across much of the historical data sets being used, and it is forensic science that is improving at a rapid rate.

  318. Lotharsson, on June 1st, 2009 at 1:54 pm Said:
    “Furthermore, you can’t point to how many hectares it burnt as a valid comparison unless you’re prepared to argue that firefighting methods were as effective then as now, that combustion loads were at similar levels per unit area, bush composition is similar to today, and perhaps even that the area of bush available to burn was similar.”

    Absolutely irrelevant!

  319. Let me flesh out your weighbridge example and see if the lightbulb goes on.

    You put about 5 tonnes on the bridge. You measure the transducer voltage as 5.605V. You add your known 1 tonne weight. You measure the transducer voltage as 6.610V.

    So you conclude that every 1.005V corresponds to one tonne.

    You put about 40 tonnes on the bridge. You measure the transducer voltage as 41.435V. You add your known one tonne weight. The transducer voltage is 42.380V. You conclude that every 0.955V corresponds to one tonne.

    So is it 0.955V or 1.005V per tonne? Or is the question itself built on fatally flawed assumptions (in this case of linearity across the full range when the transducer is non-linear)?

  320. Lotharsson, on June 1st, 2009 at 1:57 pm Said:
    “That may be true, but your procedure does not calibrate such a curve.”

    The software will.

    “If your transducer gives you a curved line, then by definition it is not full-scale linear. If that’s the case, your calibration process is fatally flawed. You don’t know WHERE on the curve your pairs of points separated by a known load lie, therefore you can’t interpolate. It’s not that hard to understand.”

    The software does.

  321. Absolutely irrelevant!

    Huh? You’re pointing to the extent of the historical fires as part of the evidence to support your case that they were worse than the recent ones. I agree this is irrelevant to climate trends, but I don’t see how it’s irrelevant to your use of that figure in your assertion.

  322. The software will.

    Now your arguments are becoming ridiculous.

    You’re merely passing the buck and then pretending that the problem doesn’t exist. Someone somewhere had to write the software, which has to solve the problem – and CAN’T because it doesn’t have the information. Look at my example from 2:12pm.

  323. Lotharsson, on June 1st, 2009 at 2:12 pm Said:
    “So is it 0.955V or 1.005V per tonne? Or is the question itself built on fatally flawed assumptions (in this case of linearity across the full range when the transducer is non-linear)?”

    It’s both, plus whatever other voltages each calibration point determines.
    I think I avoided referring to the relationship as being linear during this discussion on bridges, in fact I’m certain I did.

  324. It’s both, plus whatever other voltages each calibration point determines.
    I think I avoided referring to the relationship as being linear during this discussion on bridges, in fact I’m certain I did.

    johnd, given that you’re now explicitly talking non-linearity in transducers, what precise weight does each of the four voltages mentioned in my comment at 2:12pm represent?

  325. Lotharsson, on June 1st, 2009 at 2:15 pm Said:
    “Huh? You’re pointing to the extent of the historical fires as part of the evidence to support your case that they were worse than the recent ones.”

    In terms of area covered, yes.
    Lives lost and damage to infrastructure can be one measure, but too difficult to correlate.
    As the current enquiry reveals, once started there little that could have been done to control the fires despite all the resources available. In fact almost all resources are primarily committed to protecting small inhabited pockets that end up burnt out all round, if lucky.
    Realistically most fires run their course and it is only a change of weather that finally stops them, especially for those that rate amongst the worst.

  326. Realistically most fires run their course and it is only a change of weather that finally stops them, especially for those that rate amongst the worst.

    Even if that’s true, area burnt reveals not so much the intensity of the conditions and the sustained nature of weather conditions that are merely hot enough, and with suitably directional wind, that the fire can’t be controlled? I can agree with that, but don’t see that it makes your point about climate. It doesn’t even say that much about weather, other than the wind and rain didn’t turn on the fire for sufficiently long that it burnt out a lot of area.

    BTW, if they “run their course”, then de-bushing the landscape helps limit their course, so historically fires under the same conditions as today would probably burn much larger areas provided that they weren’t stopped by a change in weather.

    And I would suspect we have a lot more firefighting resources now that can kill off a struggling fire that back then might have been subsequently revived by an adverse change in weather.

    This line of argument seems pretty pointless.

  327. Lotharsson, on June 1st, 2009 at 2:34 pm Said:
    “johnd, given that you’re now explicitly talking non-linearity in transducers, what precise weight does each of the four voltages mentioned in my comment at 2:12pm represent?”

    Perhaps some real outputs would be better suited to pass comment on.

  328. Perhaps some real outputs would be better suited to pass comment on.

    Don’t be silly.

    The description follows your actual calibration procedure (even though the numbers are made up), and therefore there should be sufficient information to answer my question.

    Here’s another suggestion. Try and plot those four points on your calibration curve. Let me know how you go about it.

  329. Lotharsson, on June 1st, 2009 at 2:48 pm Said:
    BTW, if they “run their course”, then de-bushing the landscape helps limit their course,

    Fires generally burn through open country quicker than through scrub, then with a wind change suddenly become huge blazes.
    Actually there was an opinion within the CFA that blue gum plantations can be effective fire breaks, not sure if that still holds.
    Embers from fires can be landed long distances out in front of fires.

  330. Try and plot those four points on your calibration curve.

    … (crickets) …

    Just to make sure, I’m not asking for any interpolation or curve fitting between calibration points. I merely want to know where the four measurements that were made are on the calibration curve – or failing that, the fundamental reason why the question can’t be answered.

  331. Lotharsson, on June 1st, 2009 at 3:41 pm Said:
    “Try and plot those four points on your calibration curve. Just to make sure, I’m not asking for any interpolation or curve fitting between calibration points. I merely want to know where the four measurements that were made are on the calibration curve – or failing that, the fundamental reason why the question can’t be answered.”

    OK, here they are. Now what is the point you are trying to make?
    I assume that we both accept that these numbers nowhere near represent real output values of loadcells, so I’m interested in what point you’re going to make.
    http://nces.ed.gov/nceskids/createAgraph/default.aspx?ID=dde846eb373243c780e9365a981c9837

  332. OK, here they are. Now what is the point you are trying to make?

    Note firstly that that graph doesn’t match the described calibration procedure.

    Ask yourself – where did you get the values for your x axis for those four points? Remember that the test loads are in your calibration methodology are rough values and their precise value “does not matter’. So the test load you’ve got in the graph as a nice precise 5.0 tonnes may have actually been 5.4 or 4.8 or any of a number of other values…and the test load that you have as precisely 40.0 tonnes may have been 39.6 or 40.8 or 53.6 or any of a number of other values. You just don’t know, so you have no way of knowing what weight to report for (say) 5.930V, because the line you’ve drawn between 5 and 6 tonnes may be somewhere else entirely.

    This illustrates my point – in order to calibrate an instrument that uses piece-wise linear interpolation, you have to have accurately known test values (with known error bars!) And if you’re not using piece-wise linear interpolation, but are instead relying on full-scale linearity then it’s worse – (a) you still need an accurately known test load in order to know where your calibration point lies on the graph, plus (b) someone else has to prove that full-scale linearity is a valid characterisation of the instrument to the desired level of accuracy.

  333. Lotharsson, on June 1st, 2009 at 5:41 pm Said:
    “Note firstly that that graph doesn’t match the described calibration procedure.”

    Was it supposed to, the data you supplied was not anywhere near realistic?

    But, there is still something on the graph for you to ponder, then perhaps it will become clearer.
    For a start, the “Voltage” scale would be much smaller. Volts were only used in my earlier posts to keep it simple.
    In reality the “Y” axis (Voltage) would be scaled in mV or 4-20mA, so the range becomes quite compact relative to the “X” axis.
    On the graph drawn, it is the gradient of lines plotting the 1 tonne test weight that are only of interest. Real data would have then almost lining up with each other.
    The bridge computer stores this information and when all the test loads have been completed, the computer performs a “span” calculation which effectively connects all the dots. Remember, the bridge has been told what 1 tonne looks like at various points across it’s working range and is able to extrapolate out from these points at the appropriate gradient. It then determines a line of best fit and the bridge is then “zeroed”.
    If I remember correctly, it’s a long time ago that I had anything to do with weighbridges, the computer may tell you that the calibration is not within the required accuracy, so you run the whole load test again. The previous unknown loads are now known within a reasonable accuracy, so the 1 tonne data can be more accurately plotted on the “graph” and helps the computations achieve an even higher degree of accuracy.

  334. Was it supposed to, the data you supplied was not anywhere near realistic?

    It doesn’t matter WHAT the actual values are or even whether they are realistic, it matters that they are valid examples that illustrate the flaw in your calibration procedure.

    Remember, the bridge has been told what 1 tonne looks like at various points across it’s working range and is able to extrapolate out from these points at the appropriate gradient.

    No, no, no, no, no, no, no!

    In your procedure it has been told what an additional 1 tonne looks like at various UNKNOWN weight points across its range, therefore it is UNABLE to extrapolate out from these points at the appropriate gradient because it doesn’t know where those points are in the first place.

    Now if you substitute accurately known weights for your “rough” weights, THEN you’re in business. You have a bunch of _known_ weights with their known voltages, and you can do piecewise linear approximations (if that’s considered accurate enough) between them. Alternatively, if your manufacturer guarantees suitably linear full-range response (through a bunch of testing with KNOWN weights), then you are also in business using something like your rough load + known 1-tonne weight procedure.

    And the fact that accurate weights are required, or some other means of believing linearity applies (which itself relies on known weights), is what I’ve been saying all along.

  335. Lotharsson, on June 1st, 2009 at 7:08 pm Said:
    “therefore it is UNABLE to extrapolate out from these points at the appropriate gradient because it doesn’t know where those points are in the first place.”

    I know with smaller scales a known weight is used, but not with weigh bridges, all you need is the known test weight, and then some small, big and bigger loads.

    This is about as simple as I can make it.
    You put an unknown weight on, the bridge reads 5 volts. You add the 1 tonne test weight and the bridge reads 6 volts. You tell the bridge that the difference, 1 volt equates to 1 tonne. The light comes on and the bridge says to itself, if 1 volt equals 1 tonne, then when I read 5 volts that must have been 5 tonnes. The next weight confuses it because the 1 tonne difference only made 0.9 volt difference, but that doesn’t matter because the bridge knows how to bend the gradient to get a best fit. The bridge actually has to think laterally and connect things that don’t readily seem to fit together, whereas I sense that you are more a linear (logical) thinker. You have to try and see it from the bridges point of view.

    I can also see some resemblance here to my earlier post about BOM, El-Nino, and droughts.
    BOM see El-Nino and it automatically equals drought.
    It doesn’t occur to them that drought doesn’t always equal El-Nino.

  336. The light comes on and the bridge says to itself, if 1 volt equals 1 tonne, then when I read 5 volts that must have been 5 tonnes.

    That right there is the presumption of full-scale linearity (“1 volt = 1 tonne, therefore 5 volts = 5 tonnes”).
    And it’s either linear or you must know the actual test weights. There are no other options.

    Now a presumption of linearity is entirely plausible (and fairly common in instrumentation devices), but you have to understand (even if they didn’t mention it in the operator’s manual), that this presumption is required to make that calibration procedure valid. And that presumption can only be made if it’s backed up by someone else (i.e. the manufacturer) verifying that it’s actually true in practice.

    So I presume when you talk about “best fit” here, your calibration procedure presumes that the transducer is linear and tries to fit the best gradient to the calibration tests. That’s likely what happens with the bridge you are referring to (and probably why it flags some tests as “out of range” – it calculates they are too different given the rated error bars for the bridge to both be part of the same full-scale linear calibration curve.)

    The bridge actually has to think laterally and connect things that don’t readily seem to fit together, whereas I sense that you are more a linear (logical) thinker. You have to try and see it from the bridges point of view.

    I can think laterally, but also acknowledge the confines of logic. I think you tend to think laterally and connect things that don’t fit together, sometimes clinging to connections that don’t exist.

    And I can assure you in no uncertain terms that the bridge does NOT in any way think, let alone laterally, nor does it connect things that don’t fit together, nor does it have a point of view. I can also assure you that the bridge’s software programmer does not think laterally or connect unconnected things either. This is fairly basic physics, mathematics, and software – and no amount of lateral thinking can make something occur that violates the laws of those realms.

  337. BOM see El-Nino and it automatically equals drought.
    It doesn’t occur to them that drought doesn’t always equal El-Nino.

    The latter is entirely logical; it doesn’t require lateral thinking to make that observation.

  338. Lotharsson, on June 1st, 2009 at 10:27 pm Said:
    “And I can assure you in no uncertain terms that the bridge does NOT in any way think, let alone laterally, nor does it connect things that don’t fit together,”

    What I was alluding to, and that little graph I did helps illustrate this, is that via the software the bridge does think and reach out laterally from the information it obtained from each use of the 1 tonne test weight to connect to the next such piece of information. Because each may be of a different gradient, it then has to think how to connect them together as a best fit.
    That has nothing to do with the transducer, it just sits there just complaining louder and louder as it gets squeezed harder and harder, its main responsibility is high repeatability. It’s the bridges software that does the thinking.
    I used the term lateral to convey that idea of reaching out in that sense. I considered using fuzzy logic but I’m not sure that it was the right term either, plus it might have diverted us off on another trail altogether. 😉
    My reference to BOM was more in relation to the sentence above it regarding the bridges point of view, meaning perspective rather than lateral thought.

    PS. It’s enjoyable to discuss things like this with someone who hangs in there and takes it all in the right spirit. 🙂

  339. OK, but it’s not “lateral” and it’s not “thinking”. Using those terms confuses people who know what they mean – and who know how calibration works under the covers.

    Here what’s happening is straightforward (and deterministic) mathematical curve fitting, in this case constrained to a “curve” that is a straight line through the origin of the graph. Which implies the assumption of full-scale linearity, etc….

    (And you can improve the robustness of this curve fitting process by knowing your test weights at all times, not just your additional 1 tonne weight.)

    FWIW, note that similar (but more sophisticated) curve fitting processes are used in climate modeling to determine trends over various time intervals.

  340. So, briefly revisiting one of your climate model thoughts using the weighbridge as analogy.

    The ACME company has been using a weighbridge calibrated in the manner in which you described using their highly accurate 1 tonne weight. The calibrated measurements are accurate to +/- 2%. They weigh each truck every time it leaves the company (full) and when it arrives (empty), so they have a nice long history for each truck.

    They change over to a new weighbridge which is calibrated using the same procedure and same known 1 tonne weight, but because of technological advances is accurate to +/- 1%.

    When ACME weighs an empty truck with the new bridge, what is the precision of that measurement? Is it 2% because of the long history of weighing the same empty truck with the old bridge? Or is it 1% because the new bridge’s measurements are not affected by the old one’s measurements?

  341. Lotharsson, glad you came back, I also had some thoughts about some comments you made about how does the scale know where the accurate 1 tonne data actually sits.
    I’ve updated the graph to illustrate.
    http://nces.ed.gov/nceskids/createAgraph/default.aspx?ID=dde846eb373243c780e9365a981c9837

    Imagine the accurate 1 tonne data is the climate data we have collected over recent years, very accurate.
    The X axis is time and the Y axis is the Historical area of uncertainty.
    What is the relevance of the location of the accurate data and it’s gradient?

    With reference to your ACME weighbridge, the analogy can’t work unless the weighbridge is considered the climate, and the trucks previously loaded by shovel, are now being loaded out of a hopper with gates controlled by load cells. 🙂

  342. Lotharsson, I should have asked this question about the trucks.
    What is the range of allowable weights that the hopper gates have to work within so as not let the trucks be loaded heavier or lighter than previously?

  343. With reference to your ACME weighbridge, the analogy can’t work unless the weighbridge is considered the climate, and the trucks previously loaded by shovel, are now being loaded out of a hopper with gates controlled by load cells.

    I don’t understand why you think this is a good analogy, nor why you think the analogy to the climate can’t work.

    The weighbridge is a measuring device. It’s analogous to the methods we use to measure various climate-related variables, which have become more accurate over time. The trucks are the things we are trying to measure, and therefore analogous to the climate-related variables themselves.

  344. The X axis is time and the Y axis is the Historical area of uncertainty.
    What is the relevance of the location of the accurate data and it’s gradient?

    Sorry, but I don’t understand what you’re getting at. If you’re plotting amount of uncertainty against time, why is it increasing over time?

  345. What is the range of allowable weights that the hopper gates have to work within so as not let the trucks be loaded heavier or lighter than previously?

    I assume your point is that if you want to be 100% sure (or whatever likelihood that the manufacturer guarantees – usually less than 100%) that the trucks measured by the new weighbridge are not heavier than the trucks loaded by the old one, then they need to be (approx) 2%+1% lighter on the new weighbridge reading than on the old one, otherwise the maximum error one way on the old weighbridge and the maximum error the other way on the new could lead to a truck being actually heavier, despite reading the same.

    But despite you returning to this point over and over again, no-one, not even climatologists, would disagree. Because this is not a climate (long term averages) analogy – it’s a weather analogy (most extreme event).

  346. When ACME weighs an empty truck with the new bridge, what is the precision of that measurement? Is it 2% because of the long history of weighing the same empty truck with the old bridge? Or is it 1% because the new bridge’s measurements are not affected by the old one’s measurements?

    Can you answer this one for me? It’s really important to get this point understood and squared away.

  347. Lotharsson, on June 2nd, 2009 at 6:28 pm Said:
    “The weighbridge is a measuring device. It’s analogous to the methods we use to measure various climate-related variables, which have become more accurate over time. The trucks are the things we are trying to measure, and therefore analogous to the climate-related variables themselves.”

    You are right, actually once I thought more about it I could see that your analogy was appropriate.
    But it is not the accuracy of the weight of the trucks that is relevant, but weight of the trucks now compared to before. Are the trucks gradually getting heavier or lighter than the historical average weight measured on the old bridge.
    Lets say the trucks then averaged 100 tonnes +/- 2%, making them anywhere from 98 to 102 tonnes.
    Now with the new weighbridge, what weight do they have to be before we can declare that they are either heavier or lighter than before? 🙂

  348. Lets say the trucks then averaged 100 tonnes +/- 2%, making them anywhere from 98 to 102 tonnes.

    This is where you focus on the over-simplified case, to the detriment of understanding the scientific model. So let me answer that case first before coming back to a more sophisticated model.

    Your question is based on the simplest error model. First, a definition. An “NN% confidence interval” is the range of values within which the true value falls NN% of the time. The simplest error bar model assumes that the true value lies within the error bars every single time, so you can see that they are a 100% confidence interval.

    In that case, as I said above, for any individual measurement compared to another individual measurement, you have to (roughly) add the error bars. So if you have only one measurement of a truck on the old scale, and one of a truck (same or different) on the new, the new measurement needs to measure (slightly over) 103 tonnes before you’re 100% sure it’s heavier. (And this is the answer any climatologist would give you. It’s not violated by climate science.)

    More coming…

  349. But note what you actually asked. You asked if the trucks are heavier than the historical average weight. Now, the answer depends on how many times you weighed, and how the errors are distributed, and how confident you need to be.

    Let’s assume the errors follow a Uniform random distribution (i.e. for any given measurement attempt, they are equally likely to take on any value in the 100% confidence interval). If you weigh the same (empty) truck 50 times and average the results, what is the 100% confidence interval for that average? It’s clearly the same as the 100% confidence interval of any individual measurement. You might be supremely unlucky and get the maximum possible error every single attempt.

    But you very probably won’t. So you actually know a lot more than that.

    More coming…

  350. By virtue of averaging random errors, your average error ends up distributed more like a bell curve (formally, a “Normal distribution“, characterized by its “standard deviation“). And from that information you can figure out how wide a range you get for any level of confidence you like. So if you want to be 95% confident, then your confidence interval is approx twice the standard deviation on either side of the reported measurement; for 99.7% confident it’s approx three times the standard deviation.

    If you have enough measurements in your calculation of the average, then the actual standard deviation of your averaged errors will be pretty close to the theoretical.

    And for an error that is randomly distributed the theoretical standard deviation is 1/12th of the size of the 100% confidence interval. That means 95% of measurements of the average are within the most central 1/6th of the 100% confidence interval, and 99.7% within the most central 1/4 of it; and only about one in 16,000 measurements are outside the most central 1/3 of the 100% confidence interval.

    So if you don’t need 100% certainty (or for other reasons don’t trust your measurements in other ways that make looking for 100% pointless), then you can be far more confident in averages than you can any individual measurement.

    More coming…

  351. So it’s subtle, but important, to understand a less simplified model here. (For anyone who’s studied a science for more than one course, you’ll be familiar with the story at the start of each subsequent course – “Well, this is what we told you was true about this subject in the last course, but now we’re going to tell you the _real_ story.” And the following course starts out by saying exactly the same thing – we told you something last time, but now you’re getting the _real_ story. And so on…think Newtonian mechanics, relativistic mechanics, quantum mechanics.)

    Most real world measurement instruments are not adequately described by Uniform random distributions of errors – a Normal distribution is far more likely (and yes, that means an occasional large error is possible). And even for those instruments that have approximately Uniform random error distributions, once you start making repeated measurements for averaging or trend calculations (which in a sense average the errors associated with each individual measurement that is considered when calculating the trend), then you end up with a Normal distribution anyway. And you can answer more valuable questions, albeit with slightly less certainty, if you work with a slightly smaller confidence interval. So an awful lot of science works with Normal distributions and associated confidence intervals, rather than simple error bars with their 100% confidence.

    So the more likely and useful scientific question is – how confident are we that (say) the current truck weights are heavier than they are before? Or it is – are we NN% sure that they are heavier than they are before?

    And that’s why you see (say) the IPCC reporting on the confidence level associated with various propositions (even defining specific levels for specific English terms). These confidence levels include those that compare current climate measurements to historical climate measurements. (Figuring out the confidence interval for a comparison between two types of measurements, particularly when the error distributions change as you go back in history, is obviously more complicated than I have described, but it’s a straightforward evolution of the same type of statistics.)

    And using confidence intervals like this is standard scientific and engineering practice. Try reading a medical drug research study some time – they are never anywhere near 100% confident, but we still take the drugs the doctor prescribes. Or try asking an engineer for a 100% guarantee that your new household electronic device will work perfectly for the next 6 years and see what the response is. Or ask the engineers who designed your car, or the plane, or the train…

  352. Lotharsson, on June 2nd, 2009 at 11:37 pm Said:
    “for the next 6 years and see what the response is. Or ask the engineers who designed your car, or the plane, or the train…”

    I understand the theory of all this, but then there is the practical application. I’m reasonably familiar with the workings of a test laboratory, in particular coal test laboratories. Again this is some years ago.
    I understand the importance of adequate sampling in order to achieve a representative sample, both for exploration purposes and production.
    I understand how laboratory tests are developed to ensure a high repeatability firstly within each individual laboratory, and then amongst all laboratories, and how and why tolerances, or margins of error are developed.
    I also know how results are interpreted so as to, as much as is allowable by the built in margins of error, provide the results that are beneficial to the client, but still be explainable under scrutiny, and how environmental factors, such as humidity, can be fudged to again provide an optimistic outcome. This is an exchange that actually occurred over some disputed test results.
    “Our laboratory can’t get the same results as what you reported”
    “What was the humidity level in your lab?”
    “X”
    “Well thats why, the humidity in our lab was “Y””
    “I’ve never seen humidity levels that low in a lab before”
    “Well they were when we did the tests”

    then privately to me.
    “I really had to work hard to keep dropping the humidity until I could get the results within the specs, but if they can get it as low, then their test results will match. But I doubt it would be possible where their lab is located so they’ll just have to accept our results”

    A lot of all this may have little relevance to climate modelling. BUT irrespective of how accurate the laboratory tests are, the reliability of the results is determined when the samples are taken and prepared. Trying to get repeatability through repeated sampling can be very difficult, if not impossible if the quality of the material is not consistent.
    Once a sample has been taken and prepared for testing, repeated testing will generally yield identical results within the margins of error. If it is required to resample and then assemble a new test sample, then hopefully the new results will fall with the margins of error allowed, not by the lower margins of error of the laboratory tests, but the wider margins of error of assembling representative samples.
    The closest analogy to climate modelling is perhaps the past is represented by the exploration drilling, widely spaced holes, the intention to determine the size of the deposit and general characteristics.
    If it is determined to be viable, drill samples from more closely spaced holes are taken ahead of the mining operation so that the material when mined can be sorted into stockpiles according to quality. The sampling and testing starts to kick in.
    As the material is then loaded out for delivery the sampling and testing is then in high gear as this process provides the final test results that the contracts are based on.
    With these final samples, the highly accurate laboratory tests should find that the quality always falls within the wider range established by the production drilling, and the much broader range of the exploration drilling, though as many investors on the stock market can testify to, that doesn’t always occur.

    The point of all this narrative is to ask the question, what determines range that the final tests results should remain within before the owners start to question whether or not they are moving into a new previously unexplored deposit?

  353. Lotharsson,
    Actually coal is a good analogy. It’s real, it’s extremely variable, it was formed long ago, and it’s measurable.

    Once it’s been mined, and as I indicated, drilling ahead of the mining has narrowed down the quality in the seams being mined, it is moved into stockpiles, of generally lower, medium and higher qualities. These stockpiles are then variously drawn upon according to the current needs, perhaps even blended, and sampled and tested as they are being loaded out.
    Whilst they are drawing down the high quality stockpile, the accurate laboratory tests will determine if the average quality is within the specs for the coal which is now passing through the loadout.
    This particular average quality should, hopefully, be higher than the average quality from the medium grade stockpile, and also from the average quality across all the stockpiles.

    This then raises another point.
    If the stockpiles are built according to low, medium and high qualities, would they differ in size than if the stockpiles were built according to below average, average, and above average qualities?

  354. “Our laboratory can’t get the same results as what you reported”
    “What was the humidity level in your lab?”
    “X”
    “Well thats why, the humidity in our lab was “Y””

    That indicates that the test procedure was flawed. Humidity affected the results more than the margin of error, but the test did not specify the required humidity range for the margin of error to be valid.

    The point of all this narrative is to ask the question, what determines range that the final tests results should remain within before the owners start to question whether or not they are moving into a new previously unexplored deposit?

    I’m not sure what you’re asking – in particular I’m not clear what you’re measuring, nor how such measurements would be used to infer a transition to a previously unexplored deposit.

    If the stockpiles are built according to low, medium and high qualities, would they differ in size than if the stockpiles were built according to below average, average, and above average qualities?

    In general, yes. You’re not clear on definitions, but I would presume that “low”, “medium” and “high” were referenced to industry benchmarks, and “below average” and “above average” to the inferred average of the deposit. There would be no “average” stockpile to speak of, because “average” is a POINT in a distribution, and very few actual samples of a continuous variable will exactly match a point.

  355. Lotharsson, on June 3rd, 2009 at 4:39 pm Said:
    “I’m not sure what you’re asking – in particular I’m not clear what you’re measuring, nor how such measurements would be used to infer a transition to a previously unexplored deposit.”

    The coal we are measuring was formed millions of years ago. Like our climate it was extremely variable, and as we mine it today we find evidence of this variability.
    The exploration drilling indicated that the quality of coal within the lease being mined fell within a known range, with a number of different seams of differing qualities due to unique conditions at the time it was first formed.
    Now whilst the overall average quality of the whole deposit was determined by the exploration drilling, the highly accurate lab analysis reveals, that depending on which seam is presently being mined, the qualities of the coal may be at the high end of the range, or the low end.
    The fact that the accurately measured average quality of coal presently being mined from that particular seam is well above the overall average quality of the entire deposit, it should still be within the range of qualities determined by exploration. Even if the analysis reveals that the quality is moving outside the expected quality of the current seam, it will still be within the broader range of the entire deposit.
    However if the lab analysis reveals a quality that is outside the previously known range of the entire deposit, then perhaps the mine manager might need to consider that they are moving into an completely new area.
    However he would not make that assumption on one unusual test result. Coal being coal, and exploration drilling being limited to widely spaced holes, it would not be unusual to find pockets that yield results well outside the expected range.
    What I’m saying is that the climate data we have collected over the last couple of decades would hardly equate to one scoop from a coal sampler, let alone enough to make up a representative sample for laboratory analysis. Even all the data collected by instruments would only be equivalent to a couple of scoops.
    To do a full analysis on such a small sample would not be acceptable irrespective of how accurate the lab analysis is.
    And repeated testing of the same sample will not make it any more acceptable irrespective of of how many times it is tested or by how many different labs.

    Now, I forgotten what the question was. 🙂

  356. What I’m saying is that the climate data we have collected over the last couple of decades would hardly equate to one scoop from a coal sampler, let alone enough to make up a representative sample for laboratory analysis.

    I think that is where you go wrong by arguing from analogy.

    We’ve collected an awful lot of climate data, and not just from the last couple of decades. We compare recent climate to all of that mass of data, which is a lot like comparing your recent samples with the expectations set by your entire sampling process. And there are pretty clear trends in the last few decades that show that recent climate has been exceptional compared to the last few hundred years, if not quite a bit longer.

  357. Lotharsson, just as a coal deposit, (or any mineral deposit for that matter, gold may be an even better analogy to use) can have very small seams of coal that are interleaved between other seams, and are of exceptionally higher quality, or exceptionally lower, the deposit is evaluated as a whole and an average value determined for the entire deposit. As it is being worked through the more detailed information as to the quality of various seams is allowed for as output from them passes through the system.

    Historical climate data, that has been reconstruction from the past has come up with very wide ranging results as indicated by the band of uncertainty that becomes increasingly wider the further back in time it goes. Whilst these may be averaged out for statistical purposes, the fact that some indicators that are interleaved amongst all the others are far from the average does not mean that the indicated conditions never occurred. It may mean that they only occurred very rarely, or were of short duration.

    As with coal, an unusually high analysis, or an abnormally low result, far from the average of the particular seam, or that of the entire deposit, does not mean that there is something wrong with the sampling process, or the laboratory procedures, or that the coal is not real. It just means that the events that coincided when the coal was being formed came together for a short period of time and then reverted back to the conditions that determined the quality of the rest of the seam or deposit.

    Analysing historical data, and making projections into the future are all about number crunching. The statistics start to become more important than the reality.
    Anyone one familiar with the mining industry, and the stockmarket, would be well aware how one exceptional test result, or assay reported by an exploration company, has all the experts furiously crunching numbers and making projections as to future earnings of the latest El Dorado, but unfortunately very few of these ever match the earlier blue sky projections (except the ones I invest in 😉 )

    I also see an analogy in our legal system. The legal process is too often all about applying the letter of the law, or testing the letter of the law. All too often the original objective of seeking an obvious justice has to be discarded because the letter of the law hasn’t allowed for those particular circumstances. The lawyers make their cases using points of law. The judge, who is supposed the see the bigger picture, may do so but he is also bound to confine his judgement within certain parameters, and these often come down to the rights of one individual, rather than the rights of society as a whole, and an obvious natural justice is discarded.

Comments are closed.

%d bloggers like this: